4WS System

UNIVERSITY OF MARIBOR
FACULTY OF MECHANICAL ENGINEERING
Doctoral Thesis
SUSPENSION AND STEERING SYSTEM

DEVELOPMENT OF A FOUR WHEEL DRIVE AND
FOUR WHEEL STEERED TERRAIN VEHICLE
March, 2013 M. Sc. Shpetim LAJQI

Doctoral Thesis
SUSPENSION AND STEERING SYSTEM

DEVELOPMENT OF A FOUR WHEEL DRIVE AND
FOUR WHEEL STEERED TERRAIN VEHICLE
March, 2013 Author: M. Sc. Shpetim LAJQI

Mentor: Assoc. Prof. Dr. Stanislav PEHAN
Co-mentor: Prof. Dr. Jože FLAŠKER
University of Maribor – Faculty of Mechanical Engineering Doctoral thesis
Author: Shpetim LAJQI, M. Sc., B. Sc. Mech. Eng.

Title: Suspension and Steering System Development of a
Four Wheel Drive and Four Wheel Steered Terrain
Vehicle
UDC classification: 629.373.027.2/.3(043.3)
Keywords: terrain vehicle, suspension system, steering system, four
wheel steering
Processing of text and grafics: Shpetim LAJQI, M. Sc., B. Sc. Mech. Eng.
Proofreading: George Yeoman, English Lector, MA.
Number of copies: 11
ACKNOWLEDGEMENTS
I would like to express my gratitude and appreciation to my

mentor: Assoc. Prof. Dr. Stanislav Pehan and co-mentor:
Prof. Dr. Jože Flašker for their continuing support and guidance
throughout my years at the University of Maribor. I would also
like to thank Assoc. Prof. Dr. Mark Kegl, for his constructive
advice and effort. Special thanks to Mr. Jože Pšeničnik and his
team at “RTC” Company in Maribor, where I was part of his
project.
At the same time I would like to express my thanks to

Prof. Dr. Mujë Rugova, former rector of the University of
Prishtina and Prof. Dr. Musli Bajraktari, Dean of Faculty of
Mechanical Engineering for my selection by the academic staff
of the University of Prishtina to continue my doctoral studies at
University of Maribor as part of an agreement between the two
rectors of the University of Maribor and the University of
Prishtina.
For financial support my thanks go to the Slovene Human

Recourses Development Scholarship Fund, and the University
of Maribor.
Further thanks go to my parents and family for their dedication,

love and care. Finally, I would like to thanks my dearest wife
Mirlinda and our children Enea, Dea, and Euron, who have
always, be so patient and supportive throughout all the tough
times.
CONTAINS
1 INTRODUCTION ................................................................................ 1
1.1 Background ............................................................................................................. 1
1.2 Problem definition ................................................................................................... 2
1.2.1 Methodology for examining problem ............................................................ 4
1.3 Expected original scientific contribution .................................................................. 4
1.4 Research objectives ................................................................................................. 5
1.5 Structure of the doctoral thesis ................................................................................. 6
2 LITERATURE STUDY ........................................................................................... 8

2.1 Historical evolution ................................................................................................. 8
2.2 Function and classification of the suspension system ............................................. 11
2.2.1 The solid axle suspension system ................................................................ 12
2.2.2 The independent suspension system ............................................................ 13
2.2.3 The controllability strategies for the damping system .................................. 15
2.3 Functions and classifications of the steering systems ............................................. 16
2.3.1 Power steering system ................................................................................ 17
2.4 Four wheels steering systems................................................................................. 18
2.4.1 Mechanical four wheels steering system ..................................................... 18
2.4.2 Hydraulic four wheels steering system ........................................................ 19
2.4.3 Electro-hydraulic four wheels steering system ............................................ 19
2.4.4 Advantages of four wheels steering system ................................................. 20
2.5 Current research into the suspension and steering systems ..................................... 20
2.5.1 The current research being performed regarding suspension systems .......... 21
2.5.2 The current researches on the steering systems............................................ 23
3 RESEARCH METHODOLOGY ........................................................ 25

3.1 Literature study...................................................................................................... 25
3.2 Design processes by applying CAD systems .......................................................... 25
3.2.1 Conceptual design....................................................................................... 25
3.2.2 Preliminary design ...................................................................................... 26
3.2.3 Embodiment design .................................................................................... 27
3.2.4 Detail design ............................................................................................... 27
3.3 Numerical analyses methods .................................................................................. 28
3.4 Mathematic modeling and simulation .................................................................... 30
3.5 Optimization of the mechanical systems ................................................................ 31
3.5.1 Numerical optimization method .................................................................. 31
3.5.2 Optimization method based on a Genetic Algorithms .................................. 32
4 DEVELOPMENT OF THE SUSPENSION SYSTEM......................... 33

4.1 Motivation, design requirement, constraints and terminology................................. 33
4.1.1 Wheelbase of the vehicle l ......................................................................... 35
4.1.2 Track width of the vehicle w ...................................................................... 36
4.1.3 Sprung and un-sprung masses of the vehicle ms and mu ............................... 36
4.1.4 Design constraints and restrictions .............................................................. 36
-I-
4.2 Geometric parameters involved in the suspension system ...................................... 37

4.2.1 Suspension geometry .................................................................................. 37
4.2.2 Camber angle γ .......................................................................................... 38
4.2.3 Camber steer ............................................................................................... 40
4.2.4 Kingpin inclination angle and kingpin offset ............................................... 41
4.3 Design process for developing of the suspension system ........................................ 44
4.3.1 The first design concept .............................................................................. 44
4.3.2 The second design concept.......................................................................... 47
4.3.3 The final design concept ............................................................................. 50
4.3 Discussing of the results ........................................................................................ 54
5 MODELING, SIMULATION AND OPTIMIZATION OF THE

SUSPENSION MECHANISM FOR A TERRAIN VEHICLE .............. 55
5.1 Kinematic modeling and analyses of the suspension mechanism ............................ 56
5.1.1 Displacement analysis of the suspension mechanism .................................. 56
5.1.1.1 Vector loops and vector position ................................................... 57
5.1.1.2 Determining of the wheel path and other
characteristic of the suspension mechanism ................................... 65
5.1.1.3 Comparison of the obtained results between developed program
and the commercial Working Model Software............................... 67
5.1.2 Velocities analysis of the suspension mechanism ........................................ 69
5.1.4 Accelerations analysis of the suspension mechanism .................................. 70
5.2 Dynamic modeling of forces that acts on the suspension mechanism...................... 71
5.2.1 Dynamic equation of the free bodies ............................................................. 75
5.3 Simulation of the terrain vehicle suspension mechanism ........................................ 76
5.4 Optimization of the suspension mechanism ............................................................ 81
5.4.1 The problem of optimum design ................................................................. 81
5.4.2 The modeling of the suspension mechanism by the finite elements ............. 83
5.4.3 The optimum design of the suspension mechanism ..................................... 84
5.5 Discussing of the results ........................................................................................ 87
6 DESIGNS AND OPTIMIZATIONS OF PASSIVE, ACTIVE AND

SEMI-ACTIVE SUSPENSION SYSTEMS FOR A TERRAIN
VEHICLE.......................................................................................... 88
6.1 Mathematic modeling of the quarter vehicle suspension system ............................. 89
6.1.1 Passive linear and non-linear suspension systems........................................ 90
6.1.2 Active and semi-active non-linear suspension systems ................................ 92
6.2 Simulation of vehicle suspension by numerical methods ........................................ 94
6.3 Verifying the reliability of the developed program ................................................. 96
6.4 Optimization of the suspension parameters for passive, active, and
semi-active suspension systems ............................................................................. 98
6.4.1 Optimization of the suspension parameters for a passive linear
suspension system ...................................................................................... 98
6.4.1.1 Genetic Algorithms ....................................................................... 99
6.4.1.2 Multi-objective optimization process by using the
Genetic Algorithms ..................................................................... 101
6.4.1.3 Optimization of the passive linear suspension design parameters . 102
6.4.2 Optimization of the design parameters for the active and
semi-active non-linear suspension systems................................................ 106
- II -
6.5 Simulation of the vehicle’s suspension by using MATLAB/Simulink .................. 107

6.6 Discussion of the results ...................................................................................... 112
7 DEVELOPMENT OF THE FOUR WHEEL STEERING SYSTEM

FOR THE 4x4 TERRAIN VEHICLE ................................................ 114
7.1 Kinematics of the two/four wheels steering system .............................................. 115
7.1.1 Kinematics of the two-wheels steering system .......................................... 115
7.1.2 Kinematics of the four-wheels steering system .......................................... 117
7.2 Steering system for a terrain vehicle .................................................................... 119
7.2.1 Determining the steering wheel angles for the terrain vehicle in the
2WS mode ................................................................................................ 119
7.2.2 Determining of the steering wheel angles for the terrain vehicle in the
4WS mode ................................................................................................ 121
7.3 Design of the steering mechanism for a terrain vehicle......................................... 125
7.3.1 Design of the trapezoid steering mechanism for front wheels steering ....... 126
7.3.2 Design of the vehicle steering mechanism with 2WS by using
non-circular gears ..................................................................................... 129
7.3.2.1 Work principles of the steering mechanism with 2WS
by using non-circular gears ........................................................... 134
7.3.3 Design of the vehicle’s steering mechanism with 4WS
by using non-circular gears ....................................................................... 135
7.3.3.1 Work principles of the steering mechanism with 4WS
by using non-circular gears ........................................................... 138
7.3.4 Design of the steering mechanism with 2WS/4WS for terrain vehicle ....... 139
7.3.4.1 Work principles of the steering mechanism with 2WS/4WS .......... 139
7.4 Discussion of the results ...................................................................................... 141
8 SUMMARY AND CONCLUSION ................................................... 142

LIST OF USED SOURCES ................................................................. 145
APPENDIXES ..................................................................................... 154
Appendix A: Free body diagrams and forces that act in mechanism ............................. 155
Appendix B: Simulation and optimization of the suspension parameters for
passive, semi-active and active suspension systems for a terrain vehicle 166
Appendix C: Design of the developed terrain vehicle .................................................. 170
Appendix D: Bibliography of the candidate ................................................................. 172
Appendix E: Curriculum Vitae .................................................................................... 178
- III -
LIST OF FIGURES
Figure 1.1: Schematic presentation of the main dimensions for a new terrain vehicle .......... 3
Figure 2.1: Coach, constructed around 1650 years, exhibited in Automobile

Museum in Turin .............................................................................................. 8
Figure 2.2: Single rigid axle suspension system, invented by Elliot in 1804 years................ 9
Figure 2.3: Steering mechanism for two axes coaches; a) steering by rotation of whole front
axle, and b) steering by rotating only front wheels around the kingpin axis ....... 9
Figure 2.4: First automobile with an ICE, constructed by Karl Benz 1886 ......................... 10
Figure 2.5: Suspension and steering system for front axle.................................................. 12
Figure 2.6: A solid axle suspension system with leaf spring; a) front view,
and b) side view .............................................................................................. 12
Figure 2.7: The independent suspension system - front view ............................................. 13
Figure 2.8: Design of the independent suspension systems; a) MacPherson strut,
b) double wishbone, and c) multi-link suspension system ............................... 14
Figure 2.9: Quarter vehicle models; a) passive, b) active, and c) semi-active systems ........ 15
Figure 2.10: Mechanical front wheels steering system ......................................................... 16
Figure 2.11: Hydraulic power steering system ..................................................................... 17
Figure 2.12: Electric amplifier system ................................................................................. 17
Figure 2.13: Mechanical four wheels steering system .......................................................... 18
Figure 2.14: Hydraulic four wheels steering system ............................................................ 19
Figure 3.1: Flowchart diagram showing the stages of the design process ........................... 26
Figure 3.2: Modeling of a quarter of the suspension system by splitting into
discrete portions - FE (modeling is performed by using Pro/Engineer and
STAKx software)............................................................................................ 29
Figure 3.3: Moving capabilities of FE mapping of nodes ................................................... 29
Figure 3.4: Flowchart diagram of FEA process.................................................................. 30
Figure 3.5: Flowchart diagram of the mathematic modeling and simulation processes ....... 31
Figure 4.1: Terrain vehicle with four wheel drive and front wheel steering ........................ 34
Figure 4.2: Basic geometric dimensions of the terrain vehicle ........................................... 35
Figure 4.3: Schematic model of a quarter of the vehicle’s suspension system .................... 36
Figure 4.4: Suspension motions; a) bound, b) rest, and c) rebound .................................... 38
Figure 4.5: Vertical wheel position; a) positive, b) zero, and c) negative camber angle ...... 38
Figure 4.6: Lateral tire forces as a function of positive camber angle ................................. 39
Figure 4.7: Life expectancy of the tire as a function of camber angle ................................. 39
Figure 4.8: Positive camber steer; a) front view, and b) plan view ..................................... 40
Figure 4.9: Principle of camber steer; a) negative camber steer, and
b) positive camber steer .................................................................................. 41
Figure 4.10: Kingpin inclination angle and kingpin offset; a) side view, and b) front view .. 42
Figure 4.11: The kingpin offset cases; a) positive, b) negative, and c) zero .......................... 43
Figure 4.12: Front view of quarter double wishbone suspension system; a) longer, and
b) shorter non-equal control arms.................................................................... 45
Figure 4.13: Path of wheel for the suspension system with double wishbone
longer and shorter non-equal control arms ...................................................... 46
Figure 4.14: Vertical motion of the wheel as a function of the camber angle for suspension
system with double wishbone longer and shorter non-equal control arms ........ 47
Figure 4.15: Front view of quarter double wishbone suspension system with
longer equal control arms ................................................................................ 48
- IV -
longer non-equal length of control arm ........................................................... 49
Figure 4.17: Vertical motion of the wheel as a function of the camber angle for the
suspension system with double wishbone longer non-equal and
equal control arms .......................................................................................... 49
Figure 4.18: Final design of the 3D quarter suspension system for the terrain vehicle .......... 51
Figure 4.19: Frontal view of the 2D quarter suspension system for the terrain vehicle ......... 51
longer equal control arms and final design ...................................................... 52
Figure 4.21: Vertical motion of the wheel as a function of the camber angle
for the suspension system with double wishbones longer equal
control arms and final design ......................................................................... 53
Figure 4.22: Final design of the terrain vehicle suspension system; a) 3D view,
b) front view, and c) plane view ...................................................................... 54
Figure 5.1: Kinematic linkages of the terrain vehicles suspension mechanism ................... 56
Figure 5.2: Vector loops of four-bar links derived from the suspension mechanism ........... 57
Figure 5.3: Output angular displacement θ1 as a function of input angular displacement θ3 60
Figure 5.4: Coupler angular displacement θ2 as a function of input angular displac. θ3 ..... 61
Figure 5.5: Vector loops of three-bar links derived from the suspension mechanism.......... 62
Figure 5.6: Length of spring-shock absorber Ls as a function of input
angular displacement θ3 .................................................................................. 63
Figure 5.7: Output θ9 as a function of input angular displacement θ3 ................................. 64
Figure 5.8: Vector loops of four-bar links derived from the suspension mechanism ........... 64
Figure 5.9: Vertical motion of center of wheel ZM(θ3) as a function of input
Figure 5.10: Lateral motion of center of wheel YM(θ3) as a function of input
Figure 5.11: Path of wheel regarding the suspension mechanism ......................................... 67
Figure 5.12: Modeling of the suspension mechanism in WM environment .......................... 68
Figure 5.13: Working principles of the spring; a) compression, b) free, and c) extension ..... 71
Figure 5.14: Convention sign used for a linear spring force as a function of spring length .. 72
Figure 5.15: Convention sign used for a linear damper force as a function of piston velocity73
Figure 5.16: Free-body diagram of a vehicle during braking (longitudinal weight transfer) . 73
Figure 5.17: Simulation of ground configuration by using a hydraulic actuator ................... 76
Figure 5.18: Ground excitation zr(t) and motion of wheel ZM(t) as a function of time........... 77
Figure 5.19: Kinematic results obtained in Working Model Software .................................. 81
Figure 5.20: Modeling of terrain vehicle suspension system by Finite Elements in STAKx
software; a) side view, b) front view, c) plane view, and d) 3D view ............... 83
Figure 5.21: Analyzing of vertical wheel motion by Finite Elements in STAKx .................. 84
Figure 5.22: The optimization procedure for obtaining optimal design variables ................. 84
Figure 5.23: Optimization procedure performed in iGOx environment ................................ 85
Figure 5.24: Path of wheel for initial and optimal designs variables
of the terrain vehicle suspension mechanism ................................................... 86
Figure 6.1: The passive linear and non-linear quarter-vehicle models ................................ 90
Figure 6.2: Skyhook concept for active and semi-active non-linear suspension systems .... 92
Figure 6.3: Flowchart diagram for solving differential equations of motion ....................... 95
Figure 6.4: Ground excitation represented by double cosine road bumps ........................... 96
Figure 6.5: Laboratory devices for determining suspension characteristics ........................ 97
-V-
Figure 6.6: a) Ground excitation, and b) vertical vehicle body displacements .................... 97
Figure 6.7: GAs evaluation process ................................................................................... 99
Figure 6.8: Genetic operators (crossover and mutation) ................................................... 101
Figure 6.9: Optimization process; a) Single point between two objective functions,
and b) average distance between individuals as a function of generation ....... 104
Figure 6.10: Percentage of generation as a function of criteria met over the forth runs ...... 105
Figure 6.11: Initial and optimal values for objectives functions, as functions of vehicle
speed: a) Vehicle’s body acceleration - , and b) Dynamic tire forces - . 105
Figure 6.12: Flowchart diagram for optimization design variables for damping force ........ 106
Figure 6.13: Flowchart diagram for the simulation ground excitation, the vehicle’s
body accelerations, velocities, displacements, and tire forces as functions
of time for passive linear suspension system ................................................. 108
Figure 6.14: Flowchart diagram for simulation ground excitation, vehicle’s body
accelerations, velocities, displacements, and vertical tire forces as functions
of time for non-linear active, semi-active, and passive suspension systems ... 108
Figure 6.15: Ground excitation as a function of time ......................................................... 109
Figure 6.16: Vehicle’s body accelerations as a function of time for passive linear system . 109
Figure 6.17: Vehicle’s body velocities as a function of time for passive linear system....... 110
Figure 6.18: Vehicle’s body displacement as a function of time for passive linear system . 110
Figure 6.19: Vertical tire forces as a function of time for passive linear system ................. 110
Figure 6.20: Vehicle’s body accelerations as a function of time ........................................ 111
Figure 6.21: Vehicle’s body velocities as a function of time .............................................. 111
Figure 6.22: Vehicle’s body displacements as a function of time ....................................... 111
Figure 6.23: Vertical tire forces as a function of time ........................................................ 112
Figure 7.1: Operation modes of the 4WS; a) counter steer, b) neutral steer,
and c) same steer .......................................................................................... 114
Figure 7.2: Turning of the terrain vehicle towards the left side using the 2WS system ..... 116
Figure 7.3: Turning of the terrain vehicle towards the left side using the 4WS system ..... 117
Figure 7.4: Turning of the vehicle with 2WS; a) on the left side, and b) on the right side 120
Figure 7.5: Steering angle for the front wheels as a function of the steering wheel angle . 121
Figure 7.6: Turning of the 4WS system; a) towards the left side, and
b) towards the right side ................................................................................ 122
Figure 7.7: Steering angles for the front and rear wheels as a function of the steering
wheel angle φ when the vehicle with 4WS turns; a) towards the left side,
and b) towards the right side ......................................................................... 123
Figure 7.8: Terrain vehicle’s turning radius when operates in the 2WS and 4WS modes . 124
Figure 7.9: Vehicle turning radius as a function of steering wheel angles ϕ
for 2WS and 4WS ......................................................................................... 124
Figure 7.9: Terrain vehicle’s turning radius when operates in the 2WS and 4WS modes . 124
Figure 7.10: Road vehicle steering mechanism; a) rack and pinion, b) lever arm,
c) drag link and d) multi-link ........................................................................ 125
Figure 7.11: Trapezoidal steering mechanism; a) straight-ahead position,
b) turning towards the left side ...................................................................... 126
Figure 7.12: Comparison between functionality of the trapezoidal steering mechanism
for different β and Ackermann steering condition ......................................... 127
Figure 7.13: The error parameter e for concrete trapezoidal steering mechanism ............... 128
Figure 7.14: Cases of the bump steering; a) toe-out, b) straight-ahead direction
and b) toe-in ................................................................................................. 129
Figure 7.15: a) The non-circular gears, and b) pitch curves of the non-circular gears ......... 130
- VI -
Figure 7.16: Working position of the two pair’s non-circular gears when the
front wheels of the terrain vehicle are turning towards the: a) left side,
b) straight-ahead, and c) right side ............................................................... 131
Figure 7.17: Radiuses of the pitch curves as a function of input angular displacement δ1... 132
Figure 7.18: Angular displacements δ2(δ1) obtained from the Ackermann steering
geometry and the non-circular gears’ as a function of angular displacement .. 133
Figure 7.19: Pitch curves of the non-circular gear given by the movable coordinate system134
Figure 7.20: Steering mechanism for terrain vehicle with 2WS
by using non-circular gears ........................................................................... 135
Figure 7.21: The front and rear; a) angular displacements δ2ForR(δ1ForR) of the
driven non-circular gears, and b) pitch curves in the (xi, yi)
for driving and driven non-circular gears ...................................................... 137
Figure 7.22: Steering mechanism for the terrain vehicle with 4WS ................................... 138
Figure 7.23: Steering mechanism of the terrain vehicle with 2WS/4WS ............................ 140
Figure A.1: Free-body diagram; a) Body 1, and b) Body 2 ............................................... 155

Figure A.2: Free-body diagram; a) Body 3, b) Body 4, and c) Body 5 .............................. 156
Figure A.3: Free-body diagram for dynamic analysis of the; a) Body 6, and b) Body 7..... 157
Figure A.4: Spring, damper, spring & damper and wheel forces as a function of the time . 160
Figure A.5: Reaction forces in the joint A as a function of the time .................................. 160
Figure A.6: Reaction forces in the joint C2 as a function of the time ................................. 161
Figure A.7: Reaction forces in the joint B as a function of the time .................................. 161
Figure A.8: Reaction forces in the joint D6 as a function of the time................................. 162
Figure A.9: Reaction forces in the joint C4 as a function of the time ................................. 162
Figure A.10:Reaction forces in the joint D4 as a function of the time................................. 163
Figure A.11:Reaction forces in the joint E as a function of the time .................................. 163
Figure A.12:Reaction forces in the joint G as a function of the time .................................. 164
Figure A.13:Reaction forces in the joint I as a function of the time ................................... 164
Figure A.14:Reaction forces in the joint H as a function of the time .................................. 165
Figure B.1: Parts of the MATLAB codes for solving differential equations of
motion for passive non-linear suspension system .......................................... 166
Figure B.2: Parts of the MATLAB codes for solving differential equations of
motion for non-linear semi-active suspension system .................................... 167
Figure B.3: Configuration MATLAB/Simulink parameters for solving
the differential equations of motion............................................................... 168
Figure B.4: Windows of Multi Objective Genetic Algorithms performed in
MATLAB Optimization Tool ....................................................................... 168
Figure B.5: Flowchart diagram of the optimization process utilized in MOGAs ............... 169
Figure C.1: 2D/3D modeling of the terrain vehicle with four wheels drive and
four wheels steering performed in Pro/ENGINEER environment .................. 170
Figure C.2: 3D modeling of a front and rear suspension mechanism, engine and driver
performed in Pro/ENGINEER environment .................................................. 171
Figure C.3: 3D modeling of a front and rear suspension mechanism and driver
performed in Pro/ENGINEER environment .................................................. 171
- VII -
LIST OF TABLES
Table 1.1: Check list with main design characteristics utilized for developing of the
suspension and steering systems ......................................................................... 3
Table 4.1: Description of the used symbols for the terrain vehicle model ........................... 34
Table 4.2: Joint points and main lengths of the double wishbone suspension mechanism
with longer non-equal control arms ................................................................... 45
with shorter non-equal control arms .................................................................. 45
Table 4.4: Join point and main lengths of the double wishbone suspension mechanism
with longer equal control arms .......................................................................... 48
Table 4.5: Joint points and main length of the suspension mechanism links ....................... 52
Table 5.1: Vector representation of the four-bar linkages shown in Figure 5.2 ................... 58
Table 5.2: Vector representation of the three-bar linkages shown in Figure 5.5 .................. 63
Table 5.3: Vector representation of the four-bar linkages shown in Figure 5.8 ................... 65
Table 5.4: Comparison of the obtained results between the developed programs performed
in MathCAD and the commercial Working Model Software ............................. 68
Table 5.5: Dynamic load for different driven condition...................................................... 75
Table 5.6: Data for simulation of the suspension mechanism ............................................. 78
Table 5.7: Results comparison for angular displacement, velocity & acceleration obtained by
the developed program performed in MathCAD and Working Model Software 78
Table 5.8: Results comparison for length, relative velocity & acceleration of the
spring & shock absorber, obtained by developed program and
Working Model Software ................................................................................. 79
Table 5.9: Results comparison for total and projection in y and z-axis of velocities
for mechanism components, obtained by developed program and
Working Model Software.................................................................................. 79
Table 5.10: Results comparison for total and projection in y and z-axis of
acceleration for mechanism components, obtained by developed
program and Working Model Software ............................................................. 80
Table 5.11: Position of joints connection and design variables............................................. 85
Table 5.12: Initial and optimal values of design variables .................................................... 86
Table 6.1: Description data for road bumps when the vehicle passed over them ................. 96
Table 6.2: The MOGAs parameters performed in the MATLAB GAs toolbox................. 103
Table 6.3: Initial and optimal suspension design parameters ............................................ 104
Table 6.4: Suspension parameters used for optimization of the active damping-force ...... 107
Table 6.5: Initial and optimal design variables for the active damping force .................... 107
Table 7.1: The geometric characteristics of the steering system with 2WS ....................... 120
Table 7.2: The geometric characteristics of the steering system with 4WS ....................... 122
Table 7.3: Parameters for modeling non-circular gears for steering system with 2WS...... 132
Table 7.4: Parameters for modeling non-circular gears for steering system with 4WS...... 137
- VIII -
SUSPENSION AND STEERING SYSTEM DEVELOPMENT OF A FOUR

WHEEL DRIVE AND FOUR WHEEL STEERED TERRAIN VEHICLE
Keywords: terrain vehicle, suspension system, steering system, four wheels steering
UDC classification: 629.373.027.2/.3(043.3)
ABSTRACT
The key goal of doctoral thesis was to conceive, design, optimize, and analyze the suspension
and steering systems for a four wheel drive and four wheel steered terrain vehicle. The
common characteristic of a terrain vehicle is the greater motion of the wheels in order to
protect the vehicle from rollover risk when running over rough terrain. Known suspension
and steering systems have serious weaknesses. In order to ensure good driving comfort,
efficient driving safety, and higher maneuverability, new principles of suspension and
steering systems are proposed.
The proposed suspension system has been successfully derived from a classical double
wishbone control arm. The control arms are long but both equal. Greater wheel motion has
been ensured without reducing driving performance. In order to improve the comfort and
safety an optimal active damping force has been determined by the active and semi-active
systems. On the basis of comprehensive analysis, active system adequacy has been achieved.
The proposed suspension design provides relatively small lateral wheel motion, zero camber
angles, and it effectively absorbs the vibrations caused by ground configurations.
The goal of the developed steering system for this terrain vehicle was to design a new
steering mechanism that would provide maneuverability at low speed and suitable stability at
higher speed. This has been ensured by two modes of steering, all wheel steer and front wheel
steer. The proposed steering mechanisms conform to Ackermann steering geometry for all
modes and situations. A totally new steering concept has been developed. This effective
design consists of special pairs of gears, known as’ non-circular gears’.
- IX -
RAZVOJ OBES IN KRMILNEGA SISTEMA TERENSKEGA VOZILA S

ŠTIRIMI GNANIMI IN ŠTIRIMI KRMILJENIMI KOLESI
Ključne besede: terensko vozilo, obese sistem, krmilni sistem, štirikolesno krmiljenje
POVZETEK
Cilj doktorske disertacije je zasnovati, oblikovati, optimizirati in analizirati sistem obes in

krmilja za štirikolesno vozilo, ki ima pogon in krmiljenje na vsa kolesa. Značilnost terenskih
vozil so veliki pomiki koles, da se vozilo ne prevrne, ko se giblje po neravnem terenu. Znani
sistemi obes in krmiljenja imajo slabosti. Za zagotavljanje višje udobnosti, varnosti vozila, in
odzivnosti je treba razviti nove koncepte obes in nov krmilni sistem.
Predlagan sistem obes je učinkovita modifikacija dvojnega A-mehanizma z enakima, a
dolgima rokama, kar zagotavlja velike vertikalne pomike koles, brez poslabšanja voznih
karakteristik. Optimalna sila dušenja, ki je potrebna za izboljšanje udobja in varnosti, se
ustvarja z navideznim oziroma resničnim aktivnim sistemom, ki je dodan sitemu obes. Obširne
analize dokazujejo upravičenost uporabe aktivnih sistemov. Predlagani sistem obes
učinkovito absorbira vibracije zaradi neravnin in zagotavlja relativno majhno bočno
pomikanje koles, brez nepotrebnega nagibanja.
Cilj je razviti takšen krmilni sistem terenskega vozila, ki bi zagotavljal učinkovito
upravljanje vozila pri nizkih hitrostih in zagotavljal stabilnost pri večjih hitrostih. To je
izpolnjeno s krmiljem, ki deluje na dva različna načina. Pri enem načinu sta so aktivna vsa
štiri kolesa, pri drugem pa samo sprednja. Predlagani krmilni mehanizem z gonili in gredmi
deluje tako, da v vseh situacijah povsem sledi zakonitosti Ackermanna. Prikazan je koncept
povsem novega principa učinkovitega krmilnega sistema, ki temelji na posebnih gonilih s
specialnimi pari zobnikov.
-X-
LIST OF USED SYMBOLS
Symbol Unit Explanation of the used symbol
A, B, C, D,
Joint points of the suspension mechanism
E, F, G, H, I
A(θ3), B(θ3),
Constants for determining output angular displacement θ1
C(θ3)
D(θ3), E(θ3),
Constants for determining angular displacement θ2
F(θ3)
a12 [mm] Distance of the rotation center for the non-circular gears
-2
ai [mms ] Absolute accelerations for links of the suspension mechanism
aLs [mms-2] Relative acceleration of the spring & shock absorber (slider link)
-2
aM [mms ] Absolute acceleration at the centre of wheel
-2
aMY, aMZ [mms ] Acceleration components at the centre of wheel in y and z - axis
ax [ms-2] Longitudinal vehicle acceleration
-2
ayi, azi [mms ] Accelerations of the suspension mechanism links in yi and zi axis
b [mm] Tire width
b Design variables
bi Design variables of the suspension mechanism links
-1
bs [Nsm ] Virtual shock-absorber damping coefficient at the sprung mass
bu [Nsm-1] Virtual shock-absorber damping coefficient at the un-sprung mass
c [mm] Longitudinal distance between turning center and front axle
-1
csh [Nsm ] Shock absorber damping coefficient
csh1 [Nsm-1] Shock absorber non-linear damping coefficient
2 -2
csh2 [Ns m ] Shock absorber non-linear square damping coefficient
d [mm] Longitudinal distance between turning center and rear axle
d [mm] Tire diameter
e [deg] Error of the trapezoidal steering mechanism
2 -2
Ek [kgm s ] Kinetic energy
f(x) Multi objective function
f0(b,u) Objective function
{f} [N] Global nodal vectors of the applied forces
Fa [N] Active damping force
Fdyn-l [N] Dynamic linear force
Fdyn-n [N] Dynamic non-linear force
FFW, FRW [N] Vertical static forces acting in one tire (front or rear tires)
- XI -
FFx, FRx [N] Braking forces acting in the front and rear tires
FFz, FRz [N] Vertical forces acting in the front and rear tires
Fj [N] Reaction forces of the suspension mechanism joints
Fs [N] Spring force
Fsh [N] Shock absorber damping force
Fsh-l [N] Shock absorber linear damping force
Fsh-n [N] Shock absorber non-linear damping force
Fs-l [N] Linear spring force
Fs-n [N] Non-linear spring force
Fsta [N] Vertical static force
Ft-l [N] Dynamic linear tire force
Ft-n [N] Dynamic non-linear tire force
Fw [N] Wheel force
Fy [N] Lateral tire force
Fyj, Fzj [N] Components of the reaction forces at the ith joints; i = A … H
Fz [N] Vertical tire force
Fz-l [N] Vertical linear tire force
Fz-n [N] Vertical non-linear tire force
-2
g [ms ] Gravitational acceleration (9.81 ms-2)
H [mm] Height of the vehicle
h(x) Constraint function
H1 [m] Amplitude of the first bump
H2 [m] Amplitude of the second bump
hCG [mm] Height from ground surface of the vehicle center gravity
hj(b, u) Constraint function
i Steering gear ratio
i, j Unit vectors in the Cartesian coordinative system
i12(δ1) Gear ratio of the non-circular gears for the 2WS
i12F(δ1F) Gear ratio of the front non-circular gears for the 4WS
i12R(δ1R) Gear ratio of the rear non-circular gears for the 4WS
th
i Nodes
Jci [kg⋅m2] Mass moment inertia of the suspension mechanism links
-1
[k] [Nm ] Global stiffness matrix of a structure
-1 -1
[k] [mN ] Global inverse stiffness matrix of a structure
ks [Nm-1] Spring stiffness linear coefficient
- XII -
ks1 [Nm-1] Spring stiffness non-linear coefficient

ks2 [Nm-2] Spring stiffness non-linear square coefficient
-1
kt [Nm ] Tire stiffness linear coefficient
-1
kt1 [Nm ] Tire stiffness non-linear coefficient
kt2 [Nm-2] Tire stiffness non-linear square coefficient
-3
kt3 [Nm ] Tire stiffness non-linear cube coefficient
l [mm] Wheelbase of the vehicle
L [mm] Length of the vehicle
L1(x1, y1) Pitch curve of the non-circular gear1 for the 2WS
L2(x2, y2) Pitch curve of the non-circular gear 2 for the 2WS
lF [mm] Distance from the front axle to the vehicle center gravity
lR [mm] Distance from the rear axle to vehicle the centre gravity
Ls [mm] Spring free length
Ls(def) [mm] Spring length in compression or extension cases
m [kg] Total mass of the terrain vehicle
mi [kg] Mass of the suspension mechanism links
mn12 [mm] Module of the non-circular gears for the 2WS
mn12F [mm] Module of the front non-circular gears for the 4WS
mn12R [mm] Module of the rear non-circular gears for the 4WS
ms [kg] Vehicle’s body mass (sprung mass)
mu [kg] Wheel assembly mass (un-sprung mass)
n Number of design variables
nc Number of constraints
nobj Number of objective functions
Pe [kW] Engine power
r [mm] Trapezoid offset arms
r1(δ1) [mm] Radius of the pitch curve for the non-circular gear 1 of the 2WS
r2(δ2) [mm] Radius of the pitch curve for the non-circular gear 2 of the 2WS
r1F(δ1F) [mm] Radius of the pitch curve for the front non-circular gear 1 of the 4WS
r1R(δ1R) [mm] Radius of the pitch curve for the rear non-circular gear 1 of the 4WS
r2F(δ2F) [mm] Radius of the pitch curve for the front non-circular gear 2 of the 4WS
r2R(δ2R) [mm] Radius of the pitch curve for the rear non-circular gear 2 of the 4WS
R2WS [mm] Vehicle turning radius at the center of rear axle
R2WSmax [mm] Maximal vehicle turning radius for the 2WS mode
R2WSmin [mm] Minimal vehicle turning radius for the 2WS mode
- XIII -
R4WS [mm] Vehicle turning radius at the vehicle center gravity

R4WS [mm] Maximal vehicle turning radius for 4WS mode
R4WSmin [mm] Minimal vehicle turning radius for 4WS mode
Ri [mm] Vectors position of the suspension mechanism links (ith = 1… 9)
ri [mm] Length of the suspension mechanism links (ith = 1… 9)
s(θ3) [mm] Relative displacement of the spring & shock absorber
t [s] Time
{u} [mm] Global nodal vectors displacement
ui, vi, wi [mm] Vector displacement of ith nodes
v [kmh-1] Vehicle speed
v1 [kmh-1] Vehicle speed over the first bump
-1
v2 [kmh ] Vehicle speed over the second bump
vi [ms-1] Absolute velocities of the suspension mechanism links
vLs(t) [mms-1] Relative velocity of the spring & shock absorber
-1
vM [mms ] Absolute velocity at the centre of the wheel
-1
vMY, vMZ [mms ] Velocity components at the center of the wheel in y and z - axis
vyi, vzi [mms-1] Velocities of the suspension mechanism links in yi and zi axis
w [mm] Track width of the vehicle
W [mm] Width of the vehicle
x Design variables
x, y, z [mm] Cartesian coordinative system
x1, y1 [mm] Movable axis of the non-circular gear 1 for the 2WS
x2, y2 [mm] Movable axis of the non-circular gear 2 for the 2WS
x1F, y1F [mm] Movable axis of the front non-circular gear 1 for the 4WS
x1R, y1R [mm] Movable axis of the rear non-circular gear 1 for the 4WS
x2F, y2F [mm] Movable axis of the front non-circular gear 2 for the 4WS
x2R, y2R [mm] Movable axis of the rear non-circular gear 2 for the 4WS
xi [m] Part of road for simulation ground excitations
YE, ZE [mm] Position at the joint E
YM, ZM [mm] Position at the centre of the wheel
YN, ZN [mm] Position at the bottom of the tire
zmax [mm] Bound wheel position +250 mm
zmin [mm] Rebound wheel position -250 mm
zrest [mm] Rest wheel position 0 mm
zr [m] Ground excitation
- XIV -
zs [m] Vertical displacement of the sprung mass

zu [m] Vertical displacement of the un-sprung mass
( z s − zu ) [m] Suspension travel
(zu − z r ) [m] Tire deflection
-1
(z&s − z&u ) [ms ] Relative velocity of the spring & shock absorber
-1
z&s [ms ] Velocity of the sprung mass
-1
z&u [ms ] Velocity of the un-sprung mass
-2
&z&s [ms ] Acceleration of the sprung mass
-2
&z&u [ms ] Acceleration of the un-sprung mass
γ [deg] Camber angle
σ [deg] Kingpin inclination angle
ϕ [deg] Steering wheel angle
µ - Coefficient of friction between tire and road surface
α [deg] Road slop angle
α1 [deg] Weight coefficient corresponding to the ground excitation
β [deg] Trapezoidal offset angle
ξ [deg] Trapezoidal coupler angle
ΣFyi,ΣFzi [N] Sums of the internal and external components of force in yi and zi axis
Σ Mxi [N⋅m] Sum of all moments acting in the suspension mechanism links
δzs [m] Virtual displacement of the sprung mass
δzu [m] Virtual displacement of the un-sprung mass
δW [Nm] Elementary work
δL [deg] Front steering angle of the left wheel for the 2WS
δR [deg] Front steering angles of the right wheel for the 2WS
δLF [deg] Front steering angles of the left wheel for the 4WS
δLR [deg] Rear steer angles of the left wheel for the 4WS
δRF [deg] Front steer angles of the right wheel for the 4WS
δRR [deg] Rear steer angles of the right wheel for the 4WS
δ1 [deg] Angular displacement of the non-circular gear 1 for the 2WS
δ2(δ1) [deg] Angular displacement of the non-circular gear 2 for the 2WS
δ1F [deg] Angular displacement of the front non-circular gear 1 for the 4WS
δ1R [deg] Angular displacement of the rear non-circular gear 1 for the 4WS
δ2F(δ1F) [deg] Angular displacement of the front non-circular gear 2 for the 4WS
δ2R(δ1R) [deg] Angular displacement of the rear non-circular gear 2 for the 4WS
θ3 [deg] Input angular displacement of the 4th link
- XV -
θi(θ3) [deg] Angular displacements of the ith suspension mechanism links

θx, θy, θy [deg] Rotational vectors of ith nodes in axis x, y and z
ω1(δ1) [s-1] Angular velocity of the non-circular gear 1 for the 2WS
-1
ω2(δ2) [s ] Angular velocity of the non-circular gear 2 for the 2WS
ω1F(δ1F) [s-1] Angular velocity of the front non-circular gear 1 for the 4WS
-1
ω1R(δ1R) [s ] Angular velocity of the rear non-circular gear 1 for the 4WS
-1
ω2F(δ2F) [s ] Angular velocity of the front non-circular gear 2 for the 4WS
ω2R(δ2R) [s-1] Angular velocity of the rear non-circular gear 2 for the 4WS
-1
ωi(t) [s ] Angular velocity of the ith suspension mechanism links
ε [deg] Caster angle
εi, εi(t) [s-2] Angular acceleration of the ith suspension mechanism links
[ms-2] Standard deviation of the vehicle’s body acceleration
[N] Standard deviation of the dynamic tire forces
- XVI -
LIST OF USED ABBREVIATIONS
Abbreviations Explanation of the used abbreviations
CAD Computer Aided Design

CG Center of gravity of the vehicle
CPU Control Programming Unit
ECU Electronic Control Unit
FE Finite Element
FEA Finite Element Analyses
FEM Finite Element Methods
GAs Genetic Algorithms
GM General Motors
ICE Internal Combustion Engine
MOGAs Multi Objective Genetic Algorithms
QVM Quarter vehicle model
RTC Razvojno tehnološki center (Institute for Automotive Industry)
SAE Society Automotive Engineering
SLA Short-Long Arm
VSI Vehicle Safety Institute
2D Two dimensional view
2WD Two wheels drive
2WS Two wheels steering
4WS/2WS Four wheels steering/ Two wheels steering
3D Three dimensional view
4WD Four wheels drive
4WS Four wheels steering
4x4 Four wheel drive
- XVII -
1 INTRODUCTION
1.1 Background
The key issue for terrain vehicles is to ensure good tire contact with the ground’s surface. In
regard to specific ground roughness this can be assured by more or less comprehensive
suspension and steering systems. The common characteristic of the terrain vehicle is greater
motion of the wheels in order to protect the vehicle from rollover risk when it runs over rough
terrain. When considering this fact, the existing suspension systems has shown weakness
when the motion of the wheels is greater. In order to ensure good driving comfort, driving
safety, and higher maneuverability, it is necessary to develop new suspension systems. A
suspension system is considered ideal when it has the capability of isolating the vehicle’s
body from the vibration generated from road bumps, cornering, braking, and acceleration.
The suspension system physically connects the vehicle’s chassis with its wheels,
cushions all the ground loads to the vehicle, thus enabling the vehicle to be driven, brake, and
steered in reasonable comfort and safety. The suspension system is fixed onto the vehicle’s
chassis and consists of the wheels with tires, springs, shock absorbers, and a few rods and
linkages, as well as the steering system, Pehan et al. [1], Lajqi et al. [2 and 3].
The steering system is a decisional component for all types of vehicles. It represents a
connection between driver and vehicle, Rill [4]. The main task of it is to steer a vehicle in the
desired direction safely and under full control. Due to the importance of the steering system, it
should be easily and safely done.
The traditional steering system design is usually a steering mechanism done by some
kind of mechanical ways, Pehan et al. [1] and Rill [4]. By such a mechanism, safety is
ensured. In general, the driver not only steers the vehicle but he gets a feedback from the road
surface that is expressed by the contra torque on the steering wheel, Pehan et al. [1]. To avoid
the complicated linkage between steering wheel and steered wheels, some prototypes of
vehicles have been developed recently where the steering system is implemented by wire, [5].
At the current stage of the market the customers still highly prefer the mechanical steering
mechanism helped by electric, hydraulic or electro-hydraulic amplifier. This is because of the
safety reasons.
-1-
However, nowadays the steering mechanisms are developed enough to fulfill the
expectations for most ordinary vehicles, but when it is a question of the terrain vehicle’s
situation it is different. The standard steering mechanisms are unable to follow Ackermann
law in order to provide suitable stability as a result of greater wheels motion. Vertical greater
wheel motion does not influence the contra torque on the steering wheel, but has the ability to
rotate the steered wheels around its kingpin axles for minor angles. Thus causes the vehicle to
decrease stability as a result of the bumps steering. These already described problems are
multiplied by the greater wheels and shorter wheelbase, which is certainly characteristic of the
modern development of terrain vehicles. Four wheel drives is an additional deepening
problem of the terrain vehicle steering because of the shortage place. The power supply to the
driven wheels, on the one hand, and the steering mechanism, on the other, now would be in
the same place. The mass is complete when the four wheel drive and four wheel steering are
requested at the same time.
1.2 Problem definition
The treated problem in this research project is the development of a suspension and steering
systems for a terrain vehicle with four wheel drives and four wheel steering. The specific
requirements that are necessary to respect during the design process are that the vehicle
should provide suitable driving comfort, driving safety, and higher maneuverability. Driving
comfort is, in relation to greater vertical wheels motion, driving safety dependent on the
quality of contact by the tires with the road surface, and higher maneuverability requires big
rotations of the wheels, around its kingpin axles.
Developing of the suspension and steering system is a result of the RTC Company in
Maribor which is in the process of developing a new terrain vehicle. One of the sensibility
parts that require particular attention are identified designing a new suspension and steering
systems which should fulfill the requirement given previously. These firm requirements are as
a result a designer of the RTC Company manufacturing a prototype of the terrain vehicle.
These terrain vehicles were in operation, but during testing showed weakness in driving
comfort, stability, and maneuverability. Therefore this approach tries to improve these
weaknesses in order to design an effective suspension and steering systems. Some main
dimensions (Figure 1.1), fundamental requirements and limits about a new terrain vehicle are
presented by a check list, Table 1.1. These requirements are necessary to respect throughout
the development process and for any eventually change should be given reasonable attention.
-2-
Rear suspension
system
Front suspension
system
CG
lR
w
lF
Figure 1.1: Schematic presentation of the main dimensions for a new terrain vehicle
Table 1.1: Check list with main design characteristics utilized for developing of the
suspension and steering systems [6].
No. Characteristics of a new terrain vehicle Value Unit Remarks

1. Engine: Pe = ~ 400 [kW] Limit
2. Power transmission: Gearbox Mechanical Request
3. Vehicle speed: v= 0 ... 180 [kmh-1] Limit
4. Maximal passing road slop angle: α= ~ 50 [deg] Wish
5. Geometric dimensions:
Wheelbase: l= 2800
Track width: w = 2100
[mm] Limit
Distance: lF = 1400
Distance: lR = 1400
6. Total mass of the terrain vehicle: m= 1150 [kg] Limit
7. Dimensions of the front and rear wheels:
Tire diameter: d = 800
[mm] Request
Tire width: b = 250
8. Number of seats: n= 2 [place] Request
9. Camber angle: γ= ~0 [deg] Limit
10. Minimal vehicle turning radius: Rmin = 2.5 [m] Limit
11. Suspension system design:
Independent suspension Request
Front and rear axles:
12. Steering system design: Mechanical Request
13. Maximal vertical wheel motion:
Rebound: zmin = - 250
Rest position: zres = 0.0 [mm] Request
Bound: zmin = + 250
-3-
In Table 1.1 obviously shown that, the new suspension and steering systems require the
vertical wheel motion to be ± 250 mm, zero camber angle, and a minimal turning radius 2.5
m. All these characteristics have affects in comfort, safety, and maneuverability.
1.2.1 Methodology for examining problem
On the basis of previous experiences, it was known that, for the suspension and steering
systems, it would be impossible to satisfy all the requirements shown on the check list, Table
1.1. Therefore, the new design for both systems it is necessary.
The suspension system is developed by analyzing several designs of the suspension
mechanism. In order to evaluate performances that provides known systems, the lateral wheel
motion and camber angle are taken in consideration. Perfect results have been shown for the
proposed design, which is derived from double wishbone with long equal control arms. To
improve the comfort and safety, the active and semi-active systems were proposed. On the
basis of comprehensive analysis, it is shown that active systems provide better behavior
compared with semi-active or passive ones. The developed suspension system offers
relatively small lateral wheel motion, zero camber angle, and effectively absorbs vibrations
caused by ground excitation. Therefore, it is ensures an improvement in driving comfort and
driving safety due to better absorbs vibration and quality contact of the tires with the ground.
In order to develop a proper steering mechanism it was foreseen to design a special
steering gearbox and transmissions that would fit into each kingpin axis of the steerable
wheel. The steering gearbox consists of planetary gears, bevel gears, and several pairs of non-
circular gears that provide variables gear ratio. The non-circular gears are designed to follow
Ackermann steering geometry. The wheels’ transmission consists of a pair of bevel gears. The
driver effort is transmitted from steering gearbox into front and rear steerable wheels through
the steering propeller shafts. All the mentioned components are rarely used in steering
systems but in the case of terrain vehicles these components are necessary in order to provide
maneuverability at low speed and suitable stability at higher speed. This is ensured by two
modes such as all wheels steering and front wheels steering. The proposed steering
mechanisms follow the Ackermann steering geometry in all modes and situations.
1.3 Expected original scientific contribution
The objectives of this approach are conceiving, designing, optimizing and analyzing of the
suspension and steering systems for a four wheel drive and four wheel steered terrain vehicle.
Known solutions have shown serious weaknesses regarding common characteristics that
-4-
terrain vehicles have. In order to ensure good driving comfort, driving safety, and higher
maneuverability and new principles of suspension and steering systems are proposed.
The proposed suspension system was successfully derived at from a classic double
wishbone control arms. The control arms are long but both equal. Greater wheel motion is
ensured without reducing driving performance. To improve the comfort and safety, an optimal
active damping force is determined by employing an active or semi-active system. On the
basis of comprehensive analysis it is shown that active systems are more adequate. The
proposed suspension design provides relatively small lateral wheel motion, zero camber angle
and it effective absorbs the vibrations caused by ground configuration.
The developed steering system for the terrain vehicle provides good maneuverability at
low speed and suitable stability at higher speed. This is ensured by two modes of steering, all
wheel steering and front wheel steering. The proposed steering mechanisms follow the
Ackermann steering geometry in all modes and situations. A totally new steering concept has
been developed. Effective design is obtained by utilizing special pairs of gears, known as non-
circular gears, in order to follow Ackermann law.
1.4 Research Objectives
The objectives of this research were focused as follows:

i. Detailed analyses of requirements and limits given for the project as shown by the
check list for the terrain vehicle, in order to satisfy all the detailed tasks of the
project, as far as possible,
ii. Investigate and carefully analyze many original and new concepts regarding
suspension and steering systems,
iii. Development of completely new suspension mechanism and control strategies for
determining adequate damping force in order the terrain vehicle to provide suitable
driving comfort and driving safety when moving over various terrain,
iv. Development of a completely new steering system that would enable the steering of
all wheels/front wheels in mechanical ways. Therefore, the developed system would
provide good maneuverability at low speed and suitable stability at higher speed.
v. The development process would be accomplished by using various research
methods and optimization in order to obtain an optimal solution and
vi. Achieved results regarding suspension and steering system will be checked and
discussed in very detail.
-5-
1.5 Structure of the doctoral thesis
The Doctoral thesis is organized into eight chapters.

Chapter 1 presents the introduction about the suspension and steering systems for a
terrain vehicle, problem definition, expected original scientific contributions and research
objectives that are carried out in the doctoral thesis.
Chapter 2 presents the literature study for the suspension and steering systems. A brief
history of their development, functions, classifications, advantages and disadvantages,
controllability strategies, four wheel steering are outlines in this chapter. At the end of this
chapter the current research, related to the suspension and steering systems as well as
scientific contribution which differ from others are summarized.
Literature studies, designing by application of the CAD systems, analytical and
numerical simulation methods as well as optimization of the mechanical systems, are some of
the used research methods shown in Chapter 3.
Chapter 4 introduces the designing process regarding the development of the suspension
system. In order to fulfill the requirement about driving comfort, driving safety and
maneuverability, firstly, the design requirements, limitations and terminologies are presented.
Secondly, an introduction to the geometric parameters involved in the suspension system such
as camber angle, caster angle, kingpin inclination and kingpin offset are given. Furthermore,
several design concepts are analyzed in detail for deriving the final design of the suspension
mechanism. For each design, the obtained results, presented by diagrams for camber angles
and the lateral wheel motion as a result of the vertical wheel motion, are discussed.
In the chapter 5 the mathematical expressions for kinematic and dynamic behavior of
the developed suspension mechanism are derived. The mathematical expressions for several
suspension links of the mechanism are solved by employing numerical methods. For each
movable link of the suspension mechanism the expressions are derived by utilizing vector
loops and free body diagrams. The obtained results from the analytical approach are
compared with comprehensive commercial software the so-called Working Model Software.
The suspension mechanism behaviors are analyzed also with Finite Element Methods,
performed by the STAKx software. An optimization of the suspension mechanism links, with
the objective to obtain less lateral wheel motion, is performed by employing the iGOx
optimizer.
-6-
Chapter 6 introduces the procedures for design and optimization of the passive, active,
and semi-active linear and non-linear suspension systems of a terrain vehicle. The developed
procedure gives the ability to analyze vehicles’ suspension performances, resulting from
passive, active, or semi-active systems. The vehicle suspension system is represented by a
mathematical model, and the equations of motion are derived and solved by using
MATLAB/Simulink software. In order to verify the reliability of the derived computer
program, a comparison is made to measured experimental data. The decisional suspension
parameters for passive linear systems are optimized by utilizing the Multi-Objective Genetic
Algorithms. The parameters that determine active damping forces are optimized by using the
Hooke-Jeeves method which is based on non-linear programming. The performance of the
passive linear suspension systems is evaluated for initial and optimal design parameters. The
usefulness of the treated active and semi-active systems on an actual terrain vehicle is
presented and compared with the passive ones.
The design of the new steering system of a terrain vehicle with all wheels steer is
discussed in chapter 7. Some known steering systems are investigated and analyzed in detail
in order to find a suitable solution. The developed steering system is designed to operate by
two modes. In the first operation mode, when the vehicle moves at lower speed, the vehicle
steers all wheels thus ensuring higher maneuverability. In the second operation mode, at the
higher speed, the vehicle steers only the front wheels by locking the rear wheels thus
providing better driving safety. The mathematical expressions for determining the steering
characteristics are determined for all operation modes. The simulation result for all operation
modes are presented in suitable graphical format. Particular care is devoted to design of the
steering gearbox which provides exact Ackermann steering geometry. This is achieved by a
development of a special pairs of non-circular gears.
The summary and conclusion of the research work is presented in Chapter 8. The
obtained results regarding the development of the suspension and the steering systems are
presented and discussed. On the basis of comprehensive analysis, it is shown that the
developed suspension and steering system for terrain vehicles with four wheel drives and four
wheels steer have successfully fulfilled all the requirements given in the check list and; the
developed systems ensure driving comfort, driving safety, and good vehicle maneuverability.
Some of the additional materials pertaining to relevant results of the study, drawing,
diagrams, flowcharts, and publications relating to the doctoral thesis are enclosed in the
appendices.
-7-
2 LITERATURE STUDY
Chapter 2 presents the literature study for the suspension and steering systems. A brief history
of their development, functions, classifications, advantages and disadvantages, controllability
strategies, four wheel steering are outlines in this chapter. At the end of this chapter the
current research, related to the suspension and steering systems as well as scientific
contribution which differ from others are summarized.
2.1 Historical evolution
Transportation of passengers and goods before the advent of vehicles was horse-driven by
coach, then steam machines, and later by Internal Combustion Engines (ICE). So, suspension
function and its importance have been known since that time. To prove this fact, the
Automobile Museum in Turin exhibits an interesting example of a horse coach constructed
around 1650, which shows the existence of a suspension system. The body of the coach was
connected with a fourth leaf spring, such as shown in Figure 2.1. There is identification of any
damping device, probably the internal friction produced from the leaf springs and belts
provided the expected leaver of comfort, Genta and Morello [7].
Coach body
Leaf spring
Figure 2.1: Coach, constructed around

1650, exhibited in Automobile Museum in
Turin, Genta and Morello [7]
An Englishman Obadiah Elliot, in 1804 invented the single rigid axle suspension by
using the elliptical steel leaf spring shown in Figure 2.2. This suspension was used for horse
coaches as well as adapted for use in the first steam road vehicle, in the nineteen century,
before the invention of the internal combustion engine.
-8-
Chassis
Solid axle Leaf spring
Carriage
Figure 2.2: Single rigid axle suspension
wheel Road surface
system, invented by Elliot in 1804 years
For heavier two axes coaches destined for the transport of goods, inventors tried to find
a solution for constructing a kind of steering mechanism. Initially this was achieved by a
simple solution, where front axle rotates together with the wheels, Figure 2.3.a.
a) b)
Kinpin axis
Steering by rotate of Steering by rotaion

whole front axle only front wheels
Figure 2.3: Steering mechanism for two axes coaches; a) steering by rotation of whole front
axle, and b) steering by rotating only front wheels around kingpin axis, Dixon [8], Jazar [9]
Steering by rotation of the whole axle provides good geometric positioning of the
wheels, but the rotation axles require larger operational space. To avoid this weakness, in
1816 Längensberger in Germany patented his steering mechanism where the front axle always
stay parallel in regard to the rear axle, but each wheels rotates around its kingpin axis, as
shown in Figure 2.3.b, Dixon [8] and Jazar [9]. Such a mechanism enables each wheel to be
-9-
steered through different angles, which greatly reduces the resistance to motion during firm
turns. Later, in 1818 Rudolf Ackermann filed a British patent, Dixon [8]. This patent
describes the steering geometry law for two wheels in order for the wheels to roll correctly. A
considerable number of vehicles still use the Ackermann steering principle.
The first automobile with an ICE was a tricycle invented by Karl Benz 127 years ago,
but there is no evidence about the design of the suspension system, Figure 2.4. This maybe is
a result of this genius focusing on developing his internal combustion engine as a priority of
his work. Therefore, from 1886 a new epoch started for vehicles.
Rear solid axis ICE Tiller

with leaf spring
Steering mechanism
Powertrain
transmision
Figure 2.4: First automobile with an ICE, constructed by Karl Benz 1886, [10]
In 1893 Karl Benz obtained a German patent for a steering mechanism that followed the
Ackermann steering principle, the same steering system as La Macelle in 1878. In 1897 Benz
introduced the steering wheel for cars, a much superior system compared to the tiller. This
steering wheel was adopted by all manufacturers and was used in all ground vehicles [10].
Some years later, Gottlieb Daimler and some European vehicle manufacturers applied
coil springs for suspension systems. However, most manufacturers focused on leaf springs
that were less costly, and easier to manufacture and assemble, Mohd [11].
The first shock absorbers appeared around 1910, fitted in vehicles only by customers’
requests for sporty driving after the vehicle was sold. Many solutions were shown. Mostly of
these had friction shock absorbers which worked in extension stroke, as well as some that
worked in compression stroke. Hydraulic shock absorbers were used later.
Nowadays, several different designs of suspension and steering systems have been
introduced over past years, as a result of each designer wanting its vehicle to have the best
characteristics in terms of comfort, stability, and maneuverability.
- 10 -
2.2 Function and classification of the suspension system
The primary function of the vehicle’s suspension system is to keep quality contact between
the tires and the road surface, as well as to isolate the vehicle’s body from vibration coming
from road disturbances. The suspension system physically connects the vehicle’s chassis with
its wheels, and consists of wheels with tires, springs, shock absorbers, and a few rods and
linkages, as well as the steering system, Pehan et al. [1], Lajqi et al. [2 and 3].
According to Genta and Morello [7], Gispelle [12], and Duffy and Krick [13], the
secondary functions of the suspension system are identified as follows:
• Suspension mechanism should followed the motion of the wheels during upward
and downward motions in order to avoid undesired wheel motions such as bump
steering (toe-in or toe-out), camber angle, track variation, wheelbase variation, etc.,
• Support the weights of the chassis, vehicle body, engine, power train system, as well
as passengers with their baggage,
• Allow the wheels to go upwards and downwards, thus causing minimal motion of
the vehicle’s body,
• Ensure the steering system to keeps the wheels in correct alignment, and
• Allows for rapid cornering of the vehicle without extreme body roll.
Generally speaking, the designs of the suspension system are classified into two main
group’s, solid axle and independent suspension system, Pehan et al. [1], Lajqi et al. [3],
Gillespie [12], Eslaminasab [14]. Both systems are used on ordinary and commercial vehicles.
Figure 2.5 introduces the basic elements that constitute the independent suspension
system, used in most front axles of ordinary vehicle, such as:
• Coil spring – is used widely in independent suspension systems. The main
characteristic is spring stiffness which is related to the diameters of the wire and
coil, number of coils, and the shear modulus of the material. The coil spring stores
energy and provides flexibility during wheel motion caused by irregularity of the
road surface,
• Shock absorber – effectively dissipate energy after spring compression/extension
and damp vibration on the vehicle’s body. Shock absorber is a telescopic device
with hydraulic fluid that usually is fitted in parallel with the coil spring. Without
shock absorber, the vehicle body strongly oscillates, caused by road irregularity.
• Control arm – holds the steering knuckle, the wheel hub and moves upward and
downwards with wheels,
- 11 -
• Steering knuckle – provides bearing support for the wheel hub, bearings, and the
wheel assembly.
Shock
Coil spring
absorber
Tire
Steering Steering linkage Steering shaft
knuckle
Wheel Hydraulic
hub power steering
Control arm
Figure 2.5: Suspension and steering system for front axle, Fisher et al. [15]
2.2.1 The solid axle suspension system
The solid axle suspension system is also known as the dependent suspension system.
Connections of the left and right wheels are on the same solid axle, Figure 2.6. In this
situation, any movement of wheels will be transmitted to the opposite wheels causing them to
camber together. Solid drive axles are used usually on the rear axles of many passenger cars,
trucks and, on the front axle in many four wheel drive vehicles.
a) b) Multi - leaf
spring Chasiss
wheel tilted
Shackle
Other
Solid axle tilts Tire

Camber angle Solid
γ Road axle Road surface
bump
Figure 2.6: A solid axle suspension system with leaf spring; a) front view, and b) side view
The advantage of solid axles is considered to be the camber angle which is unaffected
by the rolling of the vehicle’s body. Therefore, it produces little camber when cornering. In
addition, wheel alignment is readily maintained, which contributes to minimizing tire wear.
The disadvantage of solid steerable axles is their susceptibility to shimmy steering vibrations,
heavy mass, etc. The better known types of solid axles are: Hotchkiss, Four link and De Dion.
- 12 -
2.2.2 The independent suspension system
The independent suspension system, allows one wheel to move upwards and downwards with
a minimum effect on the other wheels, Figure 2.7. Most passenger cars and light truck with a
front axle use this type of suspension system because it provides more space for installing the
engine, allows effective vertical wheel motion, offers better resistance to steering vibrations,
as well as providing good driving comfort. Although this suspension system has better
suspension capabilities for some vehicles, it does not satisfy, in general, the requirements for
terrain vehicles.
BOUND
Vehicle
body
REBOUND
Other wheel Wheel is pushed
not affected up by bump Road
bump
Road surface
Figure 2.7: The independent suspension system - front view
The disadvantages of the independent suspension system are considered to be the

complexities of the design and the manufacturing costs due to the greater number of parts.
Over the years, many types of independent suspension systems have been developed
such as MacPherson, double wishbone, multi-link, trailing arm, and swing axle. Many of
them have been discarded for different reasons. The MacPherson strut, double wishbone, and
multi-link suspension systems have found applications in many types of the vehicles.
The MacPherson strut suspension consists of a single control arm and a strut assembly
(spring and shock absorber) which allows the tire and wheel to move upward and downward.
The major components of the MacPherson strut are shown in Figure 2.8.a. Such
configurations can be used for front and rear axles. This suspension has a lesser number of
parts, lower mass, and providing considering smooth driving comfort. The McPherson strut,
regarding manufacturing cost, offers reasonable performance, but never matches the
requirements fulfilled by the double wishbones or multi-link suspension. Furthermore, this
type of suspension requires more vertical space and robust connection with vehicle body.
The double wishbone suspension in the United States of America is called “A-arms”,
and “Double Wishbones” in the United Kingdom, Figure 2.8.b.
- 13 -
a) b) c)
Vehicle
body
Coil Spring
Connection
Strut shock Upper to chassis
absorber control arm
assembly
Wheel upward
and downward
Lower
Control arm control arm Tire
Road surface
Figure 2.8: Design of the independent suspension system;

a) MacPherson strut, b) double wishbone, and c) multi-link suspension system
Double wishbone suspension systems are applied to luxury sedans and sports cars
because their design of elastic-kinematic parts allows for the provision of an optimum
compromise between handling and comfort, Genta and Morello [7]. The double wishbone
uses two control arms to hold the wheel during ground excitation. The upper and lower
control arms usually have non-equal lengths, from which the acronym SLA (short-long arm)
gets its name. During the design process of double wishbone careful analyses is required to
give good performance, Gillespie [12], Güler [16]. The disadvantages of double wishbone
compared with the MacPherson strut suspension is in the complexities of the design and the
manufacturing costs due to the increased number of parts.
The multi-link suspension system belongs to the group of independent systems shown in
Figure 2.8.c. It is used on both front and rear suspension systems and was derived at by
improving double wishbone. It uses three or more lateral arms and one or more longitudinal
arms that are not required to be of equal length. Recently, the multi-link suspension system
seems to have become the best independent system for vehicles because it offers the best
compromise between comfort, stability, and maneuverability, [17]. Moreover, such a
suspension system allows vehicles to obtain better performance compared with other types.
By taking into consideration all of these good things, the multi-link suspension system
naturally has a higher price for designing and manufacturing. In fact, the suspension geometry
needs to be checked carefully using design analysis software for obtaining better performance
of the vehicle. Except for the costs and higher complexity when designing and manufacturing,
this solution is considered suitable when developing such systems for terrain vehicles.
- 14 -
2.2.3 The controllability strategies for the damping system
Suspension systems regarding controllability strategies are categorized as passive, active, and
semi-active systems, Eslaminasab [14], Lajqi et al. [18] and Senthil [19], Figure 2.9.
Sensor Sensor
zs Sprung zs Sprung zs
¼ of the Sprung
vehicle mass mass (ms) mass (ms) mass (ms)
cvariable
Suspension k csh ks Fa CPU ks CPU
s
system Sensor Sensor
zu zu zu
Wheel-axle Un-sprung Un-sprung Un-sprung
assembly mass (mu ) mass (mu) mass (mu)
Tire kt ct kt ct kt ct
zr zr zr
Road
surface
a) b) c)
Figure 2.9: Quarter vehicle models; a) passive, b) active and c) semi-active systems
The passive suspension system includes, besides the mechanism, at least one of the
conventional spring and shock absorber, Figure 2.9.a. The spring has linear or nonlinear
characteristics, whilst the shock absorber exhibits a non-linear relationship between force and
relative velocity. In general, the hydraulic shock absorbers are used in most vehicles. They
work on the principle of fluid friction, Eslaminasab [14]. The damping effect in a hydraulic
shock absorber is created by fluid-flow through orifices that are small holes in the shock-
absorber’s piston. The characteristics of the spring and shock absorber are immutable and
cannot be adapted to any momentary operational condition of the vehicle. Thus the vehicle’s
performance is very limited and any improvements can only be done by the optimization of
springs’ and shock absorbers’ characteristics. Even though these suspension systems do not
fulfill all expectations regarding comfort and safety, they are widely used. For better control
under various operational conditions, an active system can be used, Figure 2.9.b.
This active suspension system, in addition to the already described components, is also
comprised of an actuator, sensors, and a control programming unit (CPU). Actually the shock
absorber is replaced by an active force actuator. The operational conditions of the vehicle are
continuously controlled by sensors that measure the velocities of the sprung and un-sprung
masses and lead it to the CPU that ensures correct impulses for the actuator, which creates the
desired active damping forces when required. To ensure adequate damping forces within
active systems, hydraulic actuators are used because they act strongly, provide sufficient
accuracy during operation, are simple in design, and have lower costs.
- 15 -
The semi-active suspension system is based on passive and active systems. This system
presents one that contains a passive shock absorber instead, and a variable shock absorber as
an active damping force that is automatically controlled by an integrated regulator. The
damping force is modulated in accordance with the operational conditions, which are
continuously controlled by sensors connected to CPU. The correct damping force can be
ensured by adjusting the orifice area within the shock absorber, by which the resistance of
fluid-flow is adapted, Wong [20]. Most of the variable shock absorbers’ work is based on the
electro-rheological or magneto-rheological fluids effects, Nekoui and Hadavi [21]; others
include servo-valves, and solenoid valves. The servo-valves are excellent devices for
changing the damping characteristics that likely work well but they are unfortunately very
expensive and complicated devices compared to solenoid. The semi-active system offer good
advantages under extreme driving conditions. When compared with the fully-active system,
the semi-active suspension system requires less energy, is cheaper, the simplest in design, and
provides other competitive performances when compared to passive systems, Lajqi et al. [18],
Karnopp et al. [22], Pajaziti [23], Lin and Kanellakopoulos [24], Yi and Song. [25], Hong et
al. [26], Turnip et al. [27].
2.3 Functions and classifications of the steering systems
The steering system consist of several components such as the steering wheel that is operated
by the driver, the steering shaft that transmits the driver’s commands to the steering gearbox,
the steering linkages are connected to the steering wheels in order to steer the vehicle along
the desired path, Figure 2.10.
Steering wheel
Steering shaft
Pinion gear Rack shaft
Steering gearbox
Kingpin axis Steering linkages
Steering wheels
Figure 2.10: Mechanical front wheels steering system
- 16 -
The steering gearbox is used to provide mechanical advantage as a result of gear ratio,
in order to help the driver to steer the vehicle more easily due to friction between the tires and
the road surface. To date many types of steering gearbox have been developed. Most of the
standard steering gearboxes are: rack and pinion, recirculation ball, worm and roller,
articulated steering, steering by wire, etc.
Rack and pinion is widely used in most common types of steering systems for ordinary
vehicles due to its low cost and easy manufacture, Figure 2.10. When the driver turns the
steering wheel, the pinion gear rotates and moves the rack shaft in the right or left direction.
Then through the steering linkages to steering the front wheels.
2.3.1 Power steering system
The power steering system has become more widely used over recent years due to the
increasing vehicle weight affecting steerable axles. In order to reduce the driver’s steering
effort, a power amplifier is applied to the existing steering systems. The most common power
amplifier is hydraulic, sometimes integrated within an electro-hydraulic device with electronic
control, to adjust steering effort according to the vehicle’s speed, Figure 2.11.
The power amplification is known as power steering. This system requires a power
steering pump attached to the engine and other necessary components. For safety reason all
power steering systems are designed in this way, the vehicle can be steered manually when
the engine is not running or if any failure occurs within the power amplifier. Most vehicle
manufacturers have started substituting hydraulic amplifiers with electric ones, Figure 2.12.
This is only for efficiency reasons, because the electric amplifier only needs to assist the
driver when the steering wheel is turned, whereas the hydraulic pump must run constantly and
the fuel consumption of the engine increases by about 2-3 %, Genta and Morello [7].
Oil tank
Low pressure
return pipe Steering wheel
Hydraulic
pump Electric
amplifier
High press.
pipe Worm gear
Steering
Steering
gearbox
Cooling shaft
device Servomotor
Figure 2.11: Hydraulic power steering Figure 2.12: Electric amplifier

system, Genta and Morello [7] system, Genta and Morello [7]
- 17 -
2.4 Four wheels steering systems
A vehicle with a four wheels steering (4WS) systems manifests itself when the driver through
steering wheel is able to turn both the front and rear wheels of the vehicle during the same
steering input. The concept for equipping vehicles with 4WS systems has been a hot topic in
vehicle dynamics since from the 1980’s, leading to the production of several vehicles
equipped with this type of technology in the early 1990’s. Since then, there has been a growth
of interest in researching and developing 4WS systems. There are many necessary reasons for
researching 4WS such as improving the maneuverability, stability of the vehicle, and
increasing driving comfort for drivers and passengers.
Even though 4WS has advantages over the front wheel steering, it is more complex and
expensive. Currently the cost of a vehicle with 4WS is more than the front wheel steered.
There are three types of 4WS systems: mechanic, hydraulic, and electro-hydraulic.
2.4.1 Mechanical four wheels steering system
Mechanical 4WS systems use twin steering gearboxes. The first one is to steer the front
wheels, whilst the second steers the rear wheels. Connection between the front and rear
steering gearboxes is performed by using propeller shafts, Figure 2.13.
Front
steering wheels
Steering Kingpin
linkages axis
Front steering
gearbox
Rack shaft
Steering
shaft
Steering
propeller shaft
Steering
wheel
Pinion gear
Rear
steering gearbox
Kingpin Steering
axis linkages
Rear
steering wheels
Figure 2.13: Mechanical four wheels steering system
The mechanical 4WS system it is insensitive to the vehicle speed, Erjavec [28].
- 18 -
2.4.2 Hydraulic four wheels steering system
The hydraulic 4WS system uses a hydraulic working principle and is one of the simplest
designs. A hydraulic 4WS system as designed for a model of vehicle produced by Mitsubishi
Motors is shown in Figure 2.14. The main weakness of this system is that the driver does not
feel feedback caused from road disturbances and, in some cases, this has a negative influence
on driving safety.
Fluid reserve tank

Front
steering pump
Front
Front power
steering arms
steering cylinder
Controle
valve Steering wheel
Rear
steering arms
Rear power
steering
cylinder
Rear
Rear
steering piston
steering pump
Figure 2.14: Hydraulic four wheels steering system, Erjavec [28].
Two way hydraulic cylinders are installed in the front and rear axles, thus making it
possible to turn the wheels towards the right and left sides. When driver turns the steering
wheel the steering hydraulic pumps in the front and rear side of the vehicle supply fluid to the
front and rear hydraulic cylinders.
2.4.3 Electro-hydraulic four wheels steering system
The electro-hydraulic 4WS is a system that combines an electronic control unit (ECU) with
hydraulic devices. This system provides the possibility for the vehicle to be sensitive
regarding adequate wheel steering angles and vehicle speed. The sensitivity with speed
provides the opportunity to optimize the vehicle's dynamic characteristics at any speed,
thereby producing enhanced stability at higher vehicle speed, and maneuverability at lower
speed. In this system, a speed sensor and the steering wheel’s angle sensor feed information to
the ECU. By processing the received information, the ECU commands the hydraulic system
to steer the rear wheels.
- 19 -
2.4.4 Advantages of four wheels steering system
The 4WS system is a serious effort on the part of automotive design engineers in order to
provide steering of the vehicle near the neutral steering. By advancing technology, 4WS
systems can operate fully electronically steer-by-wire, equal steer angles for front and rear
wheels, and sensors for monitoring the vehicle’s dynamics and adjusting the steering angles in
real time. By such systems, stability at high speed is improved and any negative effects of
road irregularities are minimized. By steering the rear wheels in the opposite direction to the
front wheels at low speed, the vehicle’s turning radius is greatly reduced. Therefore, vehicle
maneuvering on narrow roads and when parking, becomes easier, Erjavec [28].
Generally speaking, 4WS’s advantages include cornering capability, steering response,
straight-line stability, and lane changing and maneuverability at low-speed. In 2008, a
Renault Laguna Coupe, being the first European model, arrived on the market equipped with
all wheel steering, followed by BMW 7-series in 2009. These systems are controlled
electronically and therefore provide extended possibilities. Other models of vehicles with
4WS are: Chevrolet Suburban 2500, GM Concept truck, Jeep Hurricane, etc.
Although such a complex 4WS system has not been designed for massive production, a
number of prototypes with some of these technologies have been built and tested successfully.
2.5 Current research into the suspension and steering systems
Much of the current research regarding suspension and steering systems can be found in
technical papers published by the Society Automotive Engineers (SAE), particularly from
vehicle dynamics conferences. Other sources that help this approach are books, monographs,
journals for automotive engineering, proceeding paper, notes, and theses in order to collect
necessary information. Recently, research into suspension and steering systems has included a
wider array of the subject:
• Active controls strategies for the suspension and steering systems,

• Materials and manufacturing techniques regarding suspension and steering systems,
• Modeling, optimization and testing techniques in order to improve performances of
the suspension and steering systems,
• Improving known suspension and steering systems, and
• Developing new design for the suspension and steering systems.
The referred to papers in this thesis are proofing regarding the actual orientation of
researcher towards vehicle suspension and steering systems. The active control strategies for
- 20 -
suspension and steering systems display the great interest from researchers in order to
advance the driving comfort and maneuverability of a vehicle. Other researchers focus on
analyses for improving current systems and their designs, for better performance. Whilst, for
developing new designs for suspension and steering systems there are limited research
sources, probably due to the secrecy of manufacturer’s. The objective of this approach is not
to improve current systems but to design an innovative concept regarding a suspension and
steering systems for a terrain vehicle.
2.5.1 The current research being performed regarding suspension systems
Lajqi et al. [18] describe a simplified method for determining suspension parameters for
different types of passenger cars equipped with passive suspension systems. These parameters
are obtained by mathematical modeling of the quarter vehicle model and the experimental
measuring displacement of vehicle body and force actions between tires and road surfaces.
Mokhlespour et al. [29] in their research, present an optimization procedure for a double
wishbone suspension system with variable camber angle adjusting of hydraulic mechanism.
This system adjusts the camber angle within the range from -5.5 ... 5.5 degrees. Such
optimization and control of the camber angle in addition to the vehicle stability is effective in
the tires’ adhesion on road, and improves the service life of the tire when the camber is zero.
Allred [30] explores how a compliant mechanism can be used within vehicles’
suspension system. Such a mechanism has advantages regarding the questions of space
requirements, lower number of joints, rigid links, and simpler design for manufacturing and
assembly. Fatigue failure has been found to be a limiting design constraint that competes with
space and weight constraints. As the advantage of this concept is lower weight, it does not
require precise vehicle handling characteristics, and when comparing cost with performance
give reasonable results.
Heo et al. [31] have developed a method for computing kingpin geometry which is an
important design feature of vehicle suspension geometry. It determines the steering feelings of
drivers, the handling performance, and the steering stability of a vehicle. The idea of a
geometric kingpin axis has been used to design a relatively simple suspension mechanism
such as a double wishbone type, and various techniques have been developed for designing
the kingpin axes of multi-link suspensions. However, a kingpin axis for a real vehicle has
been unknown because there has been no method for measuring or computing it.
Maher [32] has identified control strategies for vehicle suspension system by measuring
operational data. Finding methods for identifying the values of suspension parameters has
- 21 -
considerable interest, in practice. For facilitating experimental suspension system data, a

quarter car suspension test rig has to be constructed. A combination of time and frequency
domain methods are used to identify suspension parameters, extract sprung mass, linear
stiffness, and non-linear damping model.
Bouzara and Richard [33] have analyzed the effects of vibrations on the driving comfort
and driving safety observed by a variation of different suspension coefficients, road
disturbances, and seating positions. In their study they developed a mathematical model for
simulating the dynamic behavior of a 3-D vehicle. By using this model, various types of non-
linear suspensions with semi-active and active systems can be investigated.
Gobbi and Mastinu [34] have used a simple linear model to derive at a number of
analytical formulas that describe the dynamic behavior of the passive suspension system.
They have introduced an optimization method, based on Multi-Objective Programming and
Monotonicity analysis featuring the best compromises between those conflicting objective
functions that indicate improved driving comfort, driving safety, and suspension travel.
Baumal et al. [35] demonstrated numerical optimization methods for partially
automating the design process. A global optimization technique such as Genetic Algorithms is
used to determine vehicle suspension parameters. The objective was to minimize the
movement of the passenger’s seat, subject to constraints representing driving safety and
suspension travel. Genetic Algorithms are also used to compare obtained results from cited
literature based on a gradient projection method.
Alkhatib et al. [36] has used the Genetic Algorithms method for optimizing the problem
of a linear vibration isolator. This method was extended to optimize a linear quarter vehicle
model. An optimal relation was found between the root mean square of the absolute
acceleration and the relative displacement. The optimal solution was obtained numerically by
Genetic Algorithms and by employing a cost function that contributes to minimizing the
absolute acceleration and displacements of the vehicle body.
Yu and Yu [37] describe a procedure for obtaining an optimal vehicle suspension design
on a quarter vehicle dynamic model. Two objective functions were used for minimizing the
acceleration of the sprung mass and suspension travel, subject to a number of constraints. In
solving this problem, the Genetic Algorithms have been used and consistently found to be
near optimal solutions within specified parameter ranges for several independent runs.
The proposed suspension system for the terrain vehicle was successfully derived at from
a classic double wishbone control arm. The control arms are long but both equal. Larger
wheel motion is ensured without reducing driving performance. To improve the comfort and
- 22 -
safety an optimal active damping force is determined by the active and semi-active systems.
Optimization of the suspension parameters for passive ones is performed by Multi Objective
Genetic Algorithms, whilst for active damping force by employing Hooke-Jeeves non-linear
programming method. On the basis of comprehensive analysis it is shown the active systems
are more adequate. The proposed suspension design provides relatively small lateral wheel
motion, zero camber angles, and effectively absorbs the vibrations caused by road excitation.
2.5.2 The current researches on the steering systems
Manufacturers producing Honda vehicles have developed a 4WS system in a mechanical way.
The turn of the rear wheels are depends from the steering of the front wheels. This system
provides two steering characteristics, Heisler [38]. One characteristic is when the vehicle
moves at high speed, the rear wheels steer over a small angle in the same direction as the front
wheels, and in the other when the vehicle’s moves at low speed and the rear wheels steer in
the opposite direction the front wheels. These characteristics were made possible from an
epicyclical gear set consisting of a fixed internally toothed annular ring gear in which a
planetary gear driven by an eccentric shaft revolves. This unit is incorporated within the rear
steering gearbox which controls the rear wheels’ turns.
Mazda manufactures of these vehicles have designed a 4WS by using a hydraulic power
unit, which is controlled electronically in accordance with the steering wheel angle and
vehicle speed. This design seems to be more complicated than the Honda 4WS. This system
incorporates suitable fail-safe for trouble-free operation. In the case of hydraulic or electronic
system failure then the rear wheels are locked and the vehicles operate as normal. For speeds
of less than 35 km/h, the rear wheels are steered in the opposite direction to the front wheels.
If the speed reaches 35 km/h, the rear wheels are turned to the straight-ahead position. More
than this speed and the rear wheels are steered in the same direction as the front wheels with
an angle limited to 50 [39].
Pehan et al. [1] have presented the modeling and simulation of an off-road vehicle with
a four wheel steering system. Research focused on developing a steering and suspension
system which provides reasonable driving comfort, driving safety, and maneuverability.
Based on their approach, a new mechanical steering mechanism has been introduced by which
the steering mechanism ensures good maneuvering for an off-road vehicle at low speed and,
on the other hand, a mechanism with satisfactory stability requirements at higher speeds.
Their conclusion is that the vehicle with 4WS produces better maneuverability than front
wheel steering for cases when the rear wheels are steered in the opposite direction to the front
- 23 -
ones. The stability of this vehicle at high speed is obtained by locking the rear wheels and
turning them into the straight-ahead position.
Alter [40] has investigated the performance benefits of four wheels steering in
passenger vehicles at driving speeds of around 50 km/h. The objective of the research was to
compare the performance of 4WS with 2WS under steady-state skidpan and transient lane-
changing conditions. Steady-state results prove to be nearly identical between the 4WS and
2WS runs of each vehicle. The transient effect of 4WS was to transfer a larger percentage of
the cornering force load onto the front tire.
Spentzas et al. [41] have reviewed some aspects of the kinematical theory regarding
four wheels steering vehicles and have presented some new results and conclusions. The
kinematic analyses of the vehicle with 4WS allowed for Ackermann’s law for vehicles with
2WS. The kinematic theory for vehicles with 2WS vehicles is a sub-case of generalized
kinematic theory for vehicles with 4WS.
Brabecc et al. [42] have developed a control system for vehicles with four wheels
steering and then building a model of the vehicle with 4WS in the scale 1:5 which allows for
the following behavior of the vehicle with 4WS. By means of the control computer in the
model, they will be able to assess several types of the algorithms for steering rear wheels.
These algorithms will be optimized according to lateral acceleration, and the yawing velocity.
Lee [43] has made a realistic application for the sensitive analysis of a four wheel
steering vehicle with complete suspension system, and with a comprehensive analytical tire
model implemented. A steering control strategy is developed for vehicle simulation to follow
a prescribed path. The simulation results using the optimal steering ratio are compared against
the results of the conventional two wheel steering based on the linear bicycle model.
Emura and Arakawa [44] have investigated a steering mechanism using elliptical gears
for a virtual two-wheeled vehicle. They have also developed and test a steering mechanism
for four wheeled robots that could be used in industry. The developed steering mechanism has
shown suitable results regarding wheeled robotic maneuverability.
The developed steering system for the terrain vehicle is a totally new concept from other
steering mechanisms. This system provides maneuverability at low speed and suitable
stability at higher speed. This is ensured by two operation modes of steering: all wheels steer,
and the front wheels steer. The proposed steering mechanism follows the Ackermann steering
principle in all modes and situations. A totally new steering concept has been developed. This
effective design consists of special pairs of gears, known as non-circular gears. Therefore, the
developed design offers higher maneuverability and ensures good driving safety.
- 24 -
3 RESEARCH METHODOLOGY
This chapter discusses the research methodologies utilized for developing a suspension and
steering system for a four wheel-drive and four wheel-steered terrain vehicles. The main
research methods employed during this thesis can be identified as literature studies; design
processes by applying CAD systems, numerical analyses methods, mathematical modeling
and simulation, optimization of the mechanical systems, as well as the comparison, analysis,
and evaluation of the obtained results. The main aim is to introduce how these methods can
be used to solve treated problems in order to fulfill, as far as possible, the predefined
requirements given by the check list.
3.1 Literature study
A significant amount of literature, scientific papers, patents, many original and innovative
concepts regarding suspension and steering systems [1-112] have been studied and analyzed
carefully. These investigations made it possible to find suitable solutions and to fulfill as
much as possible the requirements given in the check list (Table 1.1). Research started by
exploring the development history, classifications, functions, advantages and disadvantages,
controllability and configuration of suspension and steering systems, such as presented in
Chapter 2. By studying recent literature it was possible to collate new and relevant
information from what others have done within the same field, in order to find favorable
solutions for the suspension and steering systems of terrain vehicles.
3.2 Design processes by applying CAD systems
Design processing forms a major part of research and development activities. This process
covers the activities of preparing drawings, detailed technical specifications, as well as
necessary information for manufacturing products in order to satisfy customer requirements.
Generally speaking, the task of each engineer is to apply his/her scientific and
engineering knowledge to find one or more suitable solutions for specific design problems. In
addition to the basic design, it is necessary to optimize it according to requirements and
constraints; such are materials, technological possibilities, reasonably cost, legal restrictions,
and environmental and human source-related considerations.
- 25 -
The design task becomes clearer after recognition of the problem and defining the
criteria and requirements that engineers have to adopt and apply in order to design for
production. According to Pahl et al. [45], Mastinu et al. [46] the four stages of the design
process can be defined as is shown in Figure 3.1.
Definition of the problem
Conceptual design
Preliminary design
Embodiment design
Detail design
Prototype building & testing
Requirements and NO
constraints
YES
Design for production
Figure 3.1: Flowchart diagram showing the stages of the design process
3.2.1 Conceptual design
Conceptual design is the part of the design process that determines the method of solving and
treating a problem. At this stage it is necessary to identify the essential problems, establish the
functions, and find working principles for the problem. The basic solution could be reached
by examining the working principles of the problem and combining them with real
possibilities, in order to find a solution.
3.2.2 Preliminary design
Preliminary design is the second stage of the design process, also known as feasibility design,
and is considered as part of the conceptual design. The preliminary design is obtained by
refining the conceptual design according to the design specifications regarding the best choice
for the preliminary design.
- 26 -
3.2.3 Embodiment design
Embodiment design starts from preliminary layouts, selecting the most desirable preliminary
layouts, the working concepts of a technical product, and again refining and evaluating using
technical and economic criteria. Much work is then done to refine the concept. The geometry,
shapes, weights, and interactions are determined during this stage.
3.2.4 Detail design
The detail design is the final stage of the design process that combines the embodiment of
technical products with final instructions for manufacturing. This stage of design includes
detailed explanations about the drawings, dimensions, tolerances, sizes, forms, surface
properties of all the individual components, specifications of the materials, operating
procedures, manufacturing costs, etc.
The classification of the design process as four stages is not a unique one and is in fact a
general classification when considering complex mechanical systems. Nowadays, CAD
systems are employed for implementing the design phases mentioned previously. The more
up-to-date the greater those applications concerning CAD systems utilize computer software’s
such as AutoCAD, CATIA, Pro/ENGINEER, SolidWorks, etc.
The working principles of CAD systems are based on interactive computer graphics.
The users of CAD systems are designers who use data for executing commands via various
input devices of the computer, and develop images on the output device (monitor screen,
printer) for faster and easier designing. The designer with his conceptual design performs a
part of the design process, whilst the computer carries out the task most suited to its
capability, such as fast calculations, searching, sorting, routine operations, displaying on a
monitor, analyses of structural design, printing drawings, Balič [47].
The CAD system has many advantages compared with traditional manual design. It
improves the efficiency and quality of work, saves money during design, as well as the
manufacturing process. The other advantages of the CAD system could be hard to measure
but some of them are: increasing the productivity of design, easy corrections of drawings,
repetitive parts of a drawing, virtual reality simulation of the structural design, reducing
storage space, etc. In addition to advantages, CAD systems also have some drawbacks, such
as a high initial cost, requires adequate training and education of staff for efficient usage,
reduction of human resources, etc.
- 27 -
3.3 Numerical analyses methods
Nowadays, as also in past, numerical analysis through Finite Element Methods (FEM) is
significantly used in different engineering fields, such as aerospace, automotive, heavy
equipment and machinery, civil engineering, etc. Analysis by mechanical systems often
presents difficult engineering tasks because of the high complexities of their geometry on the
one hand, and the different applied loads (dynamic load, stiffness, etc.), on the other. The
behavior of these systems can be analyzed by setting the differential equations as a basic
principle of the mechanics. The solving of differential equations in many cases is impossible,
except in those cases where the geometrical, loading, and boundary conditions of the problem
are very simple, and makes possible direct integration, Mecitoglu [48]. On the other hand, the
solving of a mechanical problem where the geometry and loading condition are relatively
complex is impossible to solve by analytical methods. This problem successfully can be
solved by employing numerical methods. FEM belongs to the class of numerical methods
and is one of the more powerful methods for solving a wide-spectrum of engineering
problems.
The exact period of early development regarding FEM is very difficult to determine,
Sorić [49]. In 1943 it was discovered that the mathematician R. Courant had developed Finite
Element Analyses for obtaining approximate solutions for vibration systems, Mohd [11]. The
first publication on FEM appeared from J.H. Argyrisa in 1954. The number of publications
referring to FEM is relatively large. Between 1967 and 1999, a total of 467 books were
published regarding FEM, Mackerle [50]. According to Sorić [49] and Mackerle [50] the first
computer program based on Finite Elements was introduced between 1960 and 1970 and by
2002, 1538 FE codes had been registered.
The design task utilized by Finite Element Analysis (FEA) is to investigate the behavior
of a structure which is subjected to specific boundary conditions and loads. In order to make
an analysis of the mechanical structures using FEA, the real geometry of the structure first
needs to be split into discrete portions, which are known as Finite Elements (FE). Each FE is
formed by joining points known as nodes. FE and nodes form a grid so-called a mesh. This
mesh is programmed to contain the structural properties of selected materials which then
define how the structure will react according to certain loadings. The quarter suspension
system of the terrain vehicles performed by discrete FE is shown in Figure 3.2.
- 28 -
FE
Displacements
FE
Nodes
Rotation
Figure 3.2: Modeling of a quarter of the suspension system by splitting into discrete
portions - FE (modeling is performed by using Pro/Engineer and STAKx software)
Usually a FE is a non-linear formed by the joining of two nodes. Each node has the
capability to move within six degrees of freedom (when three are displacements and three are
rotations), Figure 3.3.
The relation between the stiffness matrixes, vector displacement, and the applied force
in nodes is determined by the global equation of FE. The global equation of FE is described
by the following expression:
[k]⋅ {u} = { f }, (3.1)
where [k] denotes the global stiffness matrix of a structure, {u} is the global nodal vectors
displacement, {f} introduces the global nodal vectors of the applied forces, ui, vi, wi are the
termed vectors displacement of nodes ith, θx, θy, θy define the rotational vectors of nodes ith in
the axes x, y and z and i = 1, 2, ... , n are the presented nodes.
θz Three dispacement (u1, v1, w1)

Node 1 ⇒
w2 Three rotations (θx, θy, θz)
4 3 2
θy
θz v2 Three dispacement (u2, v2, w2)
5 u2 Node 2 ⇒
i w1 Three rotations (θx, θy, θz)
θx
u1 v 1 θy
1 {u} = {global displacement vector} =
θx = {u1 v1 w1 u2 v2 w2 ... ui vi wi θx θy θz}T
Figure 3.3: Moving capabilities of FE mapping of nodes
Recently, an improved performance by computers has allowed the rapid development of

software packages for solving mechanical problems that are based on FEM. This software
- 29 -
enables modeling, analysis, and optimization, as well as modifications of designs for various
engineering problems over a short time. Numerical analysis, dynamic simulation, and
optimization can be performed by employing software packages such as STAKx, MATLAB,
Working Model, MathCAD, iGOx-i, etc. The tasks of each FEA software is based on FEM
formulation of a model, and the correct defining of input data. In practice, FEA software is
usually divided into three parts: pre-processing, processing, and post-processing. A
flowchart showing the organization of FEA software, is shown in Figure 3.4.
PRE-PROCESSING
Input data: geometry
of structure, material properties,
loads, boundary condition, etc.
Library of FE codes
PROCESSING
Solving problems in
mathematical modeling
form
{u} = [k ]−1 ⋅ { f }
POST-PROCESSING
Output data: graphic display, diagrams,
animation of structural behavior
Figure 3.4: Flowchart diagram of FEA process, Mecitoglu [48], Sorić [49], Roylance [51]
where [k]-1 denotes the global inverse stiffness matrix of a structure.

The pre-solver formulates the mathematical model by reading data from pre-processor, the
processor computes solution by solving global equation of FE, while in the post-processing
stage are presented obtained results and their interpreting.
3.4 Mathematic modeling and simulation
The mathematical modeling and simulation is an important process in the automotive and
other parts of engineering. They present the basic methods for designing and analyzing a
mechanical system. Basically, the mathematical modeling is performed in mathematical
language to represent and describe the existing model or a model that is going to be
constructed. During the modeling process, the researchers have to make several attempts in
order to be sure that the simulation results are suitable and accurate, Figure 3.5. In general,
researchers and engineers before the final production design perform the mathematical
- 30 -
modeling and simulation. Unfortunately, it is impossible to build an adequate model for

analyzing. Always exists Problems always exist that are impossible to model.
PROBLEM’S DATA MATHEMATICAL COMPUTER

Geometry, material SIMULATION
properties, loads, etc. MODELING Visualizing of results
VALIDATION /
VERIFICATION
Comparison of results
Figure 3.5: Flowchart diagram of the mathematical modeling and simulation processes
For that reason it is necessary to make some assumptions and modifications. Although
mathematical modeling describes just a part of the reality it can be very useful for analyses
and design. In most cases when modeling and simulating mechanical systems described using
mathematic expressions, software packages such MATLAB/Simulink, MathCAD are used.
3.5 Optimization of the mechanical systems
Optimization techniques for mechanical systems play an important role. The purpose of this is
to find the best ways of deriving maximum benefit from available resources. Over recent
years, several optimization methods have been developed (e.g. Gradient, numerical, Genetic
Algorithms, etc.). Optimization methods based on Gradient platform are widely used in
mechanical optimization systems. These methods have favorable convergence properties and
computation effort is very small. However, the optimization method based on Genetic
Algorithms is a stronger tool for finding the global optimal point without any requirement on
the gradient and the Hessian Matrix.
As discussed in Chapter 2, driving comfort, driving safety and maneuverability of the
vehicle can be achieved by designing an efficient suspension and steering system.
Optimization of these systems can be done by employing so-called iGOx optimizer software
which works in Gradient platform. While optimization of the suspension parameters can be
done by employing Multi Objective Genetic Algorithms, and the Hooke-Jeeves methods.
3.5.1 Numerical optimization methods
Numerical optimization methods that are used in engineering are usually divided into the
following three groups, Lenart [52]: conventional, shape and topology optimization methods.
- 31 -
The conventional optimization method is a numerical method and is one of the more
popular optimization methods due to its easy usage. This method is based on a modest
modification such as the geometric parameters of individual finite elements (e.g. cross-
section, thickness, etc.). In other hand, the topology optimization methods are probably the
most complex methodology for optimizing mechanical systems and when working intensively
to develop. Sometimes shape organization is used between conventional and topological
optimization methods. The key problem before shape optimization is the selection of suitable
ways for geometric parameterization. The parameterization process used for the optimization
of a suspension system is already complex enough. The main reason for this is the fact that
the conventional optimization process does not require substantial changes in preparation data
for the analysis of the structure. The situation is totally different when optimizing the shape of
the suspension mechanism or other structure. Automatically changing FE mesh requires a
different concept for preparing geometric data. In this situation, it is necessarily to have
several FE codes that have been tested and applied in various applications, and will be used
for optimizing the geometry of the suspension mechanism. In this case, the geometric data
prepared for modeling the suspension mechanism are so-called project elements obtained
from FE codes. These project elements are prepared in ways that depend on the design
variables. Design variables, within this context, can therefore be considered as a form of
parameterized FE mesh, and the whole approach is called the concept of suspension
mechanism parameterization.
3.5.2 Optimization method based on a Genetic Algorithms
The Genetic Algorithms (GAs) used for optimization of the mechanical system are stronger
tools for finding the global optimal point without any requirement on the gradient and the
Hessian Matrix. GAs are used for global search methods that are based on Darwin’s principle
of natural selection and genetic modification. GAs consists of initial population, fitness
function, selection function, crossover, and mutation.
Therefore, based on the wider number of publications and the research done regarding
GAs, it can be concluded that optimization of the suspension parameters based on the GAs
platform provides suitable results, Likaj et al. [53]. The multi-objective GAs optimization
method is employed for optimization of passive suspension parameters such as spring
stiffness, damping coefficient, tire stiffness, sprung and un-sprung masses of the terrain
vehicle.
- 32 -
4 DEVELOPMENT OF THE SUSPENSION SYSTEM
Nowadays, as well as in the past, the development of suspension systems for vehicles attracts
great interest by the designers and manufacturers of those vehicles. In general, the optimal
design of a vehicle’s suspension system has an influence on improving: driving comfort,
driving safety, maneuverability, and the service-lives of the vehicle’s components.
This chapter identifies the more important suspension geometric characteristics that
have an influence on the final design and performance, thus ensuring a reliable suspension
system. Each suspension characteristic such as wheelbase, track width, camber angles, camber
steering, kingpin inclination, and kingpin offset, is adequately explained.
Particular interest is concentrated on the camber angles, the kingpin inclination, and the
kingpin offset, which contribute towards driving safety as a result of better contact of the tire
with the road surface. In addition to this feature, consideration should also be given when
developing a suspension system to those design requirements and constraints shown on the
check list. Some conceptual designs are analyzed in detail, and after refining them a
conceptual design could be obtained for a suitable final design of the terrain vehicle’s
suspension system that fulfills the design criteria.
4.1 Motivation, design requirement, constraints and terminology
Developing a terrain vehicle is a process carried out by the RTC Company in Maribor [6].
The main sensitive parts of the terrain vehicle, such as the suspension and the steering system,
are treated here. The motivation for developing this project stems from the example given by
the RTC Company. They have designed and manufactured a terrain vehicle for sport and
recreation purposes. The design and view of it is presented in Figure 4.1. It appeared to have
some problems with stability and maneuverability. These problems could be quantified with
some available information at our disposal. The vertical motion of the wheels was insufficient
to protect the vehicle from roller risk when performing on uneven terrain. The suspension
components provided unsuitable driving comfort, and the vehicle’s maneuverability was
limited. The existing geometric dimensions provided insufficient stability, etc.
In order to avoid these weaknesses it was necessary to make a detailed scientific study
of the suspension and steering systems. On the other hand, this project for the terrain vehicle
was completely new and it was insufficient to make only small modification.
- 33 -
Rear axle Double

wishbone
Independent
Front axle suspesnion system
Figure 4.1: Terrain vehicle with four wheel drive and front wheel steering [6]
A completely new suspension and steering system needed designing in order to fulfill
the design criteria. The new designs had to avoid the problems mentioned previously. The
design requirements and constraints given by the check list in Table 1.1 had to be considered
during the development process. The newly-developed terrain vehicle’s suspension and
steering system would have to satisfy requirements relating to driving comfort, driving safety,
and maneuvering capabilities. These would ensure the terrain vehicle had better performance
and enable mass production, of course, after prototyping and testing. Figure 4.2 introduces the
basic model of the developed terrain vehicle with all its necessary geometric dimensions in
order to provide more clarity and practically during the development process, whilst Table 4.1
provides an explanation of the used symbols shown in Figure. 4.2.
Table 4.1: Descriptions of the used symbols for the terrain vehicle model
Symbols Descriptions of used symbols

CG Center of Gravity of the vehicle
hCG Height from the Center of Gravity of the vehicle to the ground
H Height of the vehicle
l Wheelbase of the vehicle
L Length of the vehicle
lF Distance from the front axle to the Center of Gravity of the vehicle
lR Distance from rear axle to the Center of Gravity of the vehicle
w Track width of the vehicle
W Width of the vehicle
- 34 -
CG
H
hCG
lF lR
l
L
CG
H
hCG
w
W
Figure 4.2: Basic geometric dimensions of the terrain vehicle
A lot of terminology exists relating to the suspension system’s design. The follow
describes only those that were important for usage during the development process.
4.1.1 Wheelbase of the vehicle l
The wheelbase of the vehicle l is the longitudinal distance between the centers of the front and
rear axles, as shown in Figure 4.2. The wheelbase of the vehicle has a greater influence on the
distributions of the longitudinal axle loads. If the wheelbase of the vehicle is longer the load
distribution is lower between the front and rear axles and provides more space for fitting a
vehicle’s device, whilst a vehicle with a shorter wheelbase provides greater load distribution
during the acceleration and braking processes, as well as during cornering situations.
A vehicle with a longer wheelbase offers more opportunity for fitting softer springs
resulting in an increased level of driving comfort. On the other hand, a vehicle with a shorter
wheelbase has more advantage regarding lower turning radius for the same steering input,
Reimpell et al. [54]. The value of the wheelbase when developing a terrain vehicle is l = 2800
mm and all four tires have outside diameter d = 800 mm by width b = 250 mm.
- 35 -
4.1.2 Track width of the vehicle w
The track width of the vehicle w is the distance between the centerlines of the right and left
wheels when viewed from the front or rear side, Figure 4.2. The track width is another
important factor. It has a direct influence on a vehicle’s cornering behavior and tendency to
roll, Reimpell et al. [54]. A vehicle with a larger track width has advantage over the smaller
lateral load distribution but, on the other, has disadvantages because it requires more lateral
motion in order to avoid obstacles when turning.
In general, a vehicle with wider track width will provide more stability than a vehicle
with narrow track width. The track width of the developed terrain vehicle is w = 2100 mm.
The front and rear wheels are along the same line. This size of track width provides more
stability when cornering and when a vehicle moves over terrains with higher slope.
4.1.3 Sprung and un-sprung masses of the vehicle ms and mu
The sprung mass of the vehicle ms is a mass which is carried by suspension springs (frame,
engine, transmission, body, etc.), whilst the un-sprung mass mu is a mass that is un supported
by a suspension spring (wheel axles, wheel hub, tires, wheel bearings, spring, dampers, etc.).
When the vehicle moves over bump or road holes, the un-sprung mass is rapidly accelerated
and must be decelerated by the suspension damping devices. Figure 4.3 shown a quarter
model of it by introducing the sprung and un-sprung masses of the vehicle.
Sprung mass
ms
Shock
Coil spring absorber
Un-sprung mass
mu
Tire
Road surface
Figure 4.3: Schematic model of a quarter of the vehicle’s suspension system
The total mass of the terrain vehicle m would be m = ms + mu = 1150 kg.
4.1.4 Design constraints and restrictions
Some of the requirements and restrictions were given by the RTC Company according to the
European Commission Directive on Design Rules. During the designing processes of the
- 36 -
suspension and steering systems consideration should be given to a consideration of the

materials’ properties, available parts, weight, cost as well as the possibilities for manufacture.
Although all of these are important for evaluating the design, compromises are often
necessary. In this case the criteria had to be addressed and a solution obtained that ensured
most of the requirements regarding the treated problem.
The basic condition during the conception and designing of the suspension and steering
system is knowledge regarding the vehicle’s main characteristics. Such characteristics are the
considered total mass, vehicle speeds, geometric dimensions, minimal turning radius,
maximal gradients of road slopes, dimensions of tires, and many other parameters that affect
the accuracy of the design. The required design data was available.
Restrictions placed limitations on adequate software, prototypes, experimental results,
small errors and deviations in design and manufacturing, the limitations of ECE regulations,
adjustments to the rear wheels, etc.
4.2 Geometric parameters involved in the suspension system
The function of the suspension system is to keep, as much as possible, the positions of the
tires relatively constant regarding the road surface throughout the motion of the suspension
mechanism. When designing it, it was very important to identify how the vertical motion of
wheel would be, in order to design a suitable suspension mechanism that would ensure such
wheel motion.
4.2.1 Suspension geometry
When speaking about the suspension geometry, this means how the un-sprung mass of the
vehicle would be connected to the sprung mass. These connections not only dictate the path of
the wheel, but also control those forces transmitted between the un-sprung and sprung masses.
The designing of a terrain vehicle’s suspension geometry is the same for both the front
and rear axles as an advanced model of the double wishbone. This advanced model could be
ranked as a multilink suspension system. The following provides more information and
details about its characteristics.
The critical criterion concerning the designing of this suspension system was the
ensuring of much higher vertical wheel motion in order for the tires to negotiate terrain
roughness. If the suspension mechanism does not provide efficient motion of the wheels then
when the wheels contact bumps or road holes, the wheels will continue their upward motion,
- 37 -
taking the chassis with them at the same high velocity, and causing greater acceleration by the
driver. The main task was to design a fully operational suspension system which would allow
efficient vertical wheel motion and thus the vehicle would provide suitable driving comfort
and driving stability during movement over various terrains. The check list required the wheel
stroke to be 500 mm, where the wheel position in bound is +250 mm - Figure 4.4.a, rest is 0
mm - Figure 4.4.b and rebound is -250 mm - Figure 4.4.c.
FRONT VIEW
Vehicle body
BOUND
REBOUND
+250 mm
REST
-250 mm
Road bump Road hole
a) b) Flat road c)
Figure 4.4: Suspension motions; a) bound, b) rest, and c) rebound
When the vehicle’s wheels move over bumps or road holes causing bound or rebound
motion, they must not come into contact with any part of the chassis or the vehicle body.
4.2.2 Camber angle γ
Some types of vehicles when moving over various terrains such are bumps and holes as well
as cornering, change the centers of the vertical positions of the wheels which is the so-called
camber angleγ. The camber angle is an angle between the vertical axis of the wheels and the
vertical axis of the vehicle when viewed from the front or rear. According to the configuration
of the suspension system, the camber angle has three possible positions such as positive
camber, zero camber, and negative camber, as shown in Figure 4.5.
+γ FRONT VIEW γ=0 -γ
CG
Vertical axis Vertical axis Road

of the vehicles of the wheel surface
a) Pozitive camber b) Zero camber c) Negative camber
Figure 4.5: Vertical wheel position; a) positive, b) zero and c) negative camber angle
The camber angle is termed positive camber +γ if the top of the wheel is further out than
the bottom, Figure 4.5, a zero camber γ = 0 is the case where the top and botton of the wheel
- 38 -
are in the same centerline as the vehicle, Figure 4.5.b. It is called negative camber -γ when the
bottom of the wheel is further out than the top, Figure 4.5.c.
Any camber angle increase will therefore reduce the stability and traction of the vehicle
as a result of a small contact of the tire with the road surface. The camber angle causes the
lateral tire forces so-called camber thrust force, as shown in Figure 4.6. If the camber angle
increases continuously, then the lateral tire force Fy also increases, which affects life-
expectancy of the tire, and the vehicle’s stability.
1000
Lateral tire force Fy [N]
Fz = 4500 [N]
750 Zero slip angle γ
Right tire
500
250
0
0 1 2 3 4 5 6 7 8 9 10
Camber angle γ [deg]
Figure 4.6: Lateral tire forces as a function of positive camber angle, Gillespie [12]
Studies have shown that zero camber leads to better life-expectancy when tire wear is
smaller. Positive camber would lead to more pronounced wear on the outer shoulder and
negative camber to more pronounced wear on the inside of the tire, Figure 4.7, Reimpell [54].
100
90
Life-expectancy [%]
80
70
60
50
Inner shoulder tire wear Outer shoulder tire wear
25
-2 -1 0 1 2
Figure 4.7: Life-expectancy of the tire as a function of camber angle, Reimpell [54]
When considering the advantages that suspension systems’ with zero camber angles
provide e.g. better contact of tire with ground surface and the life-expectancy of the tire is
higher, it was decided to design a suspension mechanism for our terrain vehicle that would
offer zero camber angles during the bound or rebound motion of the wheel.
- 39 -
4.2.3 Camber steer
When the wheels of a vehicle are cambered from the vertical centerline axis, the rolling radius
is smaller on one side than the other. The tire then forms part of a cone and tries to rotate
about its apex, as is shown in Figure 4.8.
a) +γ +γ
Opposing direction of
moving of the vehicle
b)
Direction of Direction of
tire roll tire roll
Apex
Generated cone
Nearside Straight-ahead Outside
wheel wheel wheel
Figure 4.8: Positive camber steer; a) front view, and b) plan view
If an over tire-roll, a point on the larger side of the tire moves further by more than a
point on the smaller side, this causes the wheel to deviate from the straight ahead direction
and produce camber steer. Positive camber steer will push wheels into turning away from
each other and produce toe-out, Figure 4.8, whilst negative camber on each side pushes the
wheels to turn towards each other, thus causing toe-in, Figure 4.9, Heisler [38]. Toe setting
affects three major issues regarding a vehicle’s performance, such are: tire wear, straight line
stability, and corner-entry handling characteristics. Toe-in improves the directional stability of
a car and reduces the tendency of the wheel to shimmy, Rill [4]. When the camber angle is
negative, negative camber steer assists the steering of the vehicle because the tires want to
follow the natural direction path created by the generated cone. Conversely, positive camber
angle hinders the steering of the vehicle as a result the tires being obligated to follow a forced
path which is different from the natural direction of tire roll, Figure 4.9.
- 40 -
-γ
a)
Centrifugal force
Natural direction of
Steered direction tire roll
Turning
of motion
radius
Rotation Centre
Instantaneous
Apex
Direction of
rotation
Smaller path
deviation scrub +γ
b)
Centrifugal force
Natural direction of Steered direction

tire roll of motion
Turning
Rotation Centre
Instantaneous
Apex radius
Direction of
rotation
Generated cone Larger path
deviation scrub
Figure 4.9: Principle of camber steer; a) negative camber steer, and b) positive camber steer
4.2.4 Kingpin inclination angle and kingpin offset
Kingpin geometry is also an important factor for designing vehicle suspension geometry. It
determines the steering feelings of drivers, the driving performance, and the steering stability
of the vehicle. The caster angle ε is the angle formed between the steering axis and the
vertical axis of the wheel, viewed from the side of the vehicle, Figure 4.10.a. The caster angle
for each wheel on an axle should be equal and provide steering stability, Dixon [8]. The
kingpin inclination angle σ is the angle that arises between the steering axis and vertical axis
- 41 -
of the wheel, viewed from the front of the vehicle, Figure 4.10.b. It is also known as steering
axis inclination and usually has a range from 0 to 200, Lee [43].
The kingpin angle helps offer space for fitting brakes. If the kingpin inclination is
determined correctly, then driver steering effort will be reduced by providing drivers with
good feeling. If the kingpin inclination angle is zero, it is called centerline steering. The
kingpin inclination axis is usually neither vertical nor centered on the tire contact patch for
reasonable issues regarding the steering stability of the vehicle.
a) SIDE VIEW b) FRONT VIEW

Caster angle ε Inclination angle σ
Kingpin axis Kingpin axis
Wheel centre
Forward (Steering axis) (Steering axis)
Direction upper joins upper joins
z z
lower joins lower joins
x y
o o
+ +
Mechanical Trail Kingpin offset
Figure 4.10: Kingpin inclination angle and kingpin offset; a) side view, and b) front view
The kingpin offset is the horizontal distance measured from the steering axis to the
intersection point with the ground and wheel center contact with ground as shown in Figure
4.10 and Figure 4.11. The kingpin offset is also known as the scrub radius which can be
positive, negative and zero offset, Figure 4.11. The kingpin offset is positive when the contact
center of the wheel is outside the steering axis intersection point on the ground, Figure 4.11.a.
If the contact center of the wheel is inside the steering axis intersection point to the ground
then the kingpin offset is negative, Figure 4.11.b. It is zero when the contact center of the
wheel is at the same point of steering axis intersection on the ground, Figure 4.11.c. The
larger kingpin inclination angles result with a smaller or negative kingpin offset.
When a driver accelerates or stops a vehicle then the vehicle’s wheels will produce
driving or braking forces (at the ground). These forces are transmitted through the steering
mechanism’s additional steering torques to the steering wheel, which is proportional to the
offset distance. If the driving or braking force is different on the left than on the right wheels
then there will be a net steering torque felt by the driver (assuming that the steering system
has good enough reverse efficiency). When the wheels have a tendency to stop or accelerate
during the braking or acceleration processes, the positive kingpin offset distance and the
- 42 -
inertia force of the vehicle produce a turning moment which force the wheels to turn about the
contact patch center in an outward direction at the front, Figure 4.11.a. Therefore the positive
kingpin offset compounds the natural tendency for the vehicle to deviate towards the left if
the right wheel slips instead of continuing on a stable straight ahead path, Heisler [38].
FRONT VIEW
a) b) c)
Uper control Kingpin axis σ
arm z (Steering axis)
y
O
Steering Lower Road
knuckle control arm surface
Positive kingpin offset Negative kingpin offset Zero offset
Wheel centre
PLANE VIEW
Inertia force Inertia force Inertia force
x
Turning moment
y
O
Contact Contact Contact
patch Pivot patch patch
centre
Ground Ground Ground
reaction force reaction force reaction force
Figure 4.11: The kingpin offset cases; a) positive, b) negative, and b) zero
With negative kingpin offset the inertia force of the vehicle will produce a turning
moment that forces the wheels to turn inwards at the front about the contact path center,
Figure 4.11.b. The consequence of negative kingpin offset is that the effective braked wheels
rotate in the opposite direction to which the vehicle tends to turn and so counteracts the
deviation tendency, enabling the vehicle to stay in a stable straight ahead direction.
Sometimes a negative offset is used, thus giving straighter braking when the surface friction
varies between tracks, Dixon [8].
The only case when turning moment (additional steering torque) will not be produced
from the wheels of the vehicle will be when the kingpin offset is zero, because there is no
turning moment, Figure 4.11.c. When developing a suspension system for a terrain vehicle the
tendency is to design the kingpin offset to be zero.
- 43 -
4.3 Designing process for developing of the suspension system
After a detailed study of the literature and by taking into consideration the advantages and
disadvantages concerning various suspension systems such as respecting the criteria and
constrains, sufficient knowledge had been developed to effectively design a suspension
system for the terrain vehicle. It was found that the best response to fulfilling the design task
was to install an independent suspension system. This independent suspension system would
make it possible for each wheel to have independent motion.
In order for the terrain vehicle to provide good performance a suspension system would
be required that would allow greater wheel motion in a vertical direction, whilst in passenger
cars such movements are considerably smaller due to the rollover risk. In order to ensure
greater motion of the wheel, the suspension system should continuously maintain contact
between tires and ground surface. Based on previous experience these requirements could be
obtained if the suspension system provides a zero camber angle. It was foreseen that an
adequate solution would be a sophistication of the double wishbone suspension system.
The following were the examined kinematics of the double wishbone suspension
systems, where the lengths of the control arms will be both equal and unequal. The developed
design is derived from double wishbone. For analyzing the suspension performance when a
vehicle passed over bumps, a comparison was carried out of the behavior of the wheel motion
when wheels moved from their static positions (rest positions – 0) to their upward limits
(bound positions +250 mm) and to their downward limits (rebound positions +250 mm).
Many authors within this research area have only analyzed the quarter vehicle
suspension system. Therefore, it seemed to be sufficient to analyze only the quarter of it. This
justification was a result of the fact that the terrain vehicle has the same suspension
mechanism on all its wheels.
4.3.1 The first design concept
The first design concept was started by analyzing the double wishbone with the unequal
control arms introduced in Figure 4.12. Figure 4.12.a shows a frontal view of the double
wishbone suspension system with equal longer arms, whilst Figure 4.12.b show unequal
shorter control arms. If the vehicle has large track width w it is better to starts from the
premise that the longer control arms of the double wishbone are better than the shorter. The
double wishbone suspension system expect their current popularity, longer control arms allow
a relatively large vertical motion of wheel which corresponding by small changes of its
- 44 -
characteristics. The suspension system with shorter control arms works reasonably well
around a limited range of motions and after that drastically changes its characteristics when it
is near the end of its motion.
FRONT VIEW
F F'
Upper control arm Spring & shock
absorber assembly
of the vehicles
Vertical axis
z
REBOUND BOUND
C C' Vehicle
A A' body
G G'
E O
D D' -y
B B'
E'
Lower control arm
H H' Tire
a) Road surface b)
w/2 w/2
Figure 4.12: Front view of the quarter double wishbone suspension system;
a) longer and b) shorter non-equal control arm
In order to confirm the statement that the suspension system with longer control arms
has better kinematics behavior compared with the shorter ones, the following necessary data
for the modeling of suspension systems with longer and shorter non-equal control arms, is
presented in Tables 4.2 and 4.3.
with longer non-equal control arms
Joint positions and length of links [mm]

Axis
A B C D E F G H

yi -323 -273 -767 -849 -561 -500 -992.5 -992.5
500 650 270 325
zi 85 -85 -126 -386 -236 300 -250 -650
The data presented in Table 4.2 is used for modeling the suspension system with longer
non-equal arms, whilst Table 4.3 introduces data for modeling the suspension system with
shorter non-equal upper and lower control arms.
with shorter non-equal control arms

Axis
A’ B’ C’ D’ E’ F’ G’ H’ ′ ′
′′ ′′
′′
yi -450 -400 -755 -826 -613 -500 -990 -990
350 500 270 250
zi 85 -85 -87 -347 -216 300 -250 -650
- 45 -
Relative motions of the wheels can be in: longitudinal x, lateral y, and vertical z
directions. During the braking and acceleration processes of the vehicle in the longitudinal
direction, caster angle and mechanical trail appear. Lateral and vertical motions of the wheels
occur when the vehicle moves over uneven terrain by producing bound and rebound positions.
Researches so far have indicated that for kinematic analyses of the wheels motion it is
suitable to consider only the vertical and lateral motions. For comparison, kinematic
advantages that provide the suspension system with double wishbone longer and shorter non-
equal control arms are simulated in Working Model environments, the results are shown by
the diagrams.
Figure 4.13 present the path of the wheel for the suspension system with double
wishbone longer and shorter non-equal lengths of the control arms. It is indicated that during
the vertical motion of the wheel in rebound for -220 mm and rest positions 0, the path of the
wheel has the same lateral displacement by intersection at the same point.
300
Vertical motion of wheel ZM [mm]
200
BOUND
Shorter non-equal control arms

Longer non-equal control arms
ZM
100
REST
0 YM
-100
REBOUND
w/2
-200 LATERAL MOTION
-300
-1050 -1040 -1030 -1020 -1010 -1000 -990 -980 -970
Lateral motion of wheel YM [mm]
longer and shorter non-equal control arms
Figure 4.14 introduces how the camber angle for longer and shorter non-equal control
arms changes as a result of motion of the wheel from its rest position. It is observed that when
the wheel moves upward from the rest position, the camber angles drastically increase by
producing negative camber angle, whilst when the wheel moves downward the camber angle
becomes more positive with smaller values than the upward motion. Also the suspension
- 46 -
system with shorter control arms produces more camber angle compare with the longer
control arms. The higher values for the camber angles are a result of the upper and lower
control arms not having equal lengths. Higher values of camber angles have an influence on
vehicle stability as a result of poor contact of the tires with the road surface.
300
Shorter non-equal control arms
200
BOUND
100
REST
-100 w/2
REBOUND
-200
-300
-14 -12 -10 -8 -6 -4 -2 0 2 4 6 8
Figure 4.14: Vertical motion of wheel as a function of camber angle for the suspension
system with double wishbone longer and shorter non-equal control arms
It can obviously be seen that the suspension system with double wishbone shorter
control arms has lower performances compared with the longer control arm. This confirms
when with wheel motion in the boundary positions, the suspension system with shorter non-
equal control arms obtains greater displacements and camber angle, and consequently
influences the life expectancy of the tires, and vehicle stability. Therefore, in the second
design concept and the final design, the suspesnion system of the terrain vehicle was designed
a longer length of control arm. The only significant disadvantage of the double wishbone
suspesnion system with longer control arms is their relatively low stiffness. This potential
problem can be overcome by using a slightly larger tube with a thicker wall.
4.3.2 The second design concept
The second design concept was an advancement of the first design concept, Figure 4.15. So,
the design of the suspension system in this case was based on the fact that the double
wishbone with longer control arms offered better features compared to the shorter. This
argument was proven during the first design concept. What differentiates from the first design
- 47 -
concept is that the upper and lower control arms have equal lengths. The comparison
reference is used double wishbone with longer non-equal control arms. Figure 4.15 introduces
the frontal view of the quarter double wishbone suspension system with longer equal control
arms. Table 4.4 gives the necessary data for modeling the suspension system with longer
equal upper and lower control arms.
F Spring & shock

Upper control arm absorber assembly
REBOUND BOUND
C z
A
G
D
E O
B -y
of the vehicles
Vertical axis
Lower control arm
H Tire
Road surface
w/2
Figure 4.15: Front view of quarter double wishbone suspension system

with longer equal control arms
Table 4.4 Joint points and main lengths of the double wishbone suspension mechanism
with longer equal control arms
Axis
A B C D E F G H

yi -273 -313 -816 -856 -584 -500 -992.5 -992.5
600 600 175 300
zi 85 -85 -171 -342 -213 300 -250 -650
The data presented in Table 4.4 will be used for modeling the suspension system with
longer equal control arms within the Working Model environments. The results are presented
in the following diagrams.
Figure 4.16 shows the path of wheel for the double wishbone suspension system with
longer non-equal and equal control arms. It is shown that during rebound the motions of the
wheel curves seem to be similar to the non-equal and equal control arms. For the rest position,
both systems have the same wheel path. However, when the wheel is in bound motion, the
difference the between the non-equal and equal control arms is very small. In other words this
small difference in lateral wheel displacement cannot be considered either as an advantage or
disadvantage that provides the suspension system with longer non-equal and equal control
arms.
- 48 -
300

200 Longer non-equal control arms
BOUND
Longer equal control arms
ZM
100
REST
0 YM
-100
REBOUND
w/2
-200 LATERAL MOTION
-300
-1050 -1040 -1030 -1020 -1010 -1000 -990
longer non-equal and equal control arms
Figure 4.17 introduces how the camber angle for longer non-equal and equal control
arms changes as a result of the motion of wheel from its rest position. It can be observed that
when the wheel moves upward or downward from the rest position, the camber angle obtain
higher values for the double wishbone with longer non-equal control arms. The situation is
totally different for the double wishbone with longer equal control arms where the change of
the camber angle from rest position is approximately zero.
300
Vertical motion ofTwheel ZM [mm]
200
BOUND
100
RES
-100 w/2
REBOUND
-200
-300
-10 -8 -6 -4 -2 0 2 4 6
suspension system with double wishbone longer non-equal and equal control arms
- 49 -
As a result of the characteristics shown in Figures 4.16 and 4.17 for the suspension
system with longer equal control arms, it can be concluded that this suspension system had
the better advantages when compared to the longer non-equal control arms. Even though it
can be considered that such a suspension system offers better performance that others, in the
boundary position the wheel (bound and rebound) from the rest position still has a large
displacements of the wheel in the lateral direction of around 58 mm left and 58 mm right for
the wheel, a total of 116 mm. Such a displacement would be expected to affect the life-
expectancy regarding driving stability by causing additional loads on the suspension
mechanism. Therefore, in the final design of the suspension system, the development of it
focused on the sophistication of the second concept design but with a new design to fulfill the
requirement that camber angle and lateral wheel motion should have smaller values as far as
possible.
4.3.3 The final design concept
The state-of-the-art for final design concept of the terrain vehicle’s suspension system is
based on three fundamental principles:
• a suitable design for robust performance,
• using the best proven performance provided by the reviewed suspension, and
• continuous refining of the design until it is optimized for fulfilling the submitted
requirement.
Actually, the advent the personal computer made it possible to analyze the suspension
geometry without cutting and welding its structures. By applying the available knowledge for
using computer simulations, it was possible to attain the predefined requirements. The only
thing still required to do was to decide what the suspension system needed to function well
and then run the computer to proceed with the variables and obtain the desired design.
The final design of the suspension system was derived at by refining the double
wishbone suspension system with a longer equal control arm, Figure 4.18.
This suspension system is categorized as a multi-link suspension system that is rather
more complicated than other systems, and 3D of quarter of it is shown in Figure 4.18. This
design allows for relatively large vertical motion of wheel by using correspondingly small
undesired suspension characteristics. The further development process focused on the lateral
motion of the tire and camber angle having, as far as possible, smaller values for the boundary
positions when the wheel motion is in bound and rebound. The design of the quarter
suspension system in 2D (y and z - axis) with necessary parameters, is shown in Figure 4.19.
- 50 -
Spring and
shock absorber
Support carriage
Vehicle
body
z
Upper control
arm -y O
Lower
control arm
Driving x
Steering axis
direction
w/2
Figure 4.18: Final design of the 3D quarter suspension system for the terrain vehicle
Design of the quarter suspension system in 2D (y and z - axis) with necessary

parameters is shown in Figure 4.19.
Spring and F z
shock absorber
Support carriage A
B Vehicle
C
G body
Knuckle
θ3
E
J H D
BOUND Upper
M L control arm -y O
Lower
REBOUND
K I control arm
Kinpin axis
N w/2
Figure 4.19: Front view of 2D quarter suspension system for terrain vehicle
Point O is centre of coordinative system -yOz with coordinates O (0, 0). Axis -yO go
through centre of the wheel (rest position), while Oz is centers of vehicle-front view. For
evaluation of the performance that provides this suspension system is necessary to make
geometric modeling. The joints position and other necessary data for geometry modeling are
given in Table 4.5. This data help to analyze performance that provides that system.
- 51 -
Table 4.5: Joint points and main lengths of the suspension mechanism links

Axis
A B C D E F G H

yi -273 -273 -351 -351 -517 -554 - 708 -876
220 220 170 170
zi 317 147 111 -59 19 501 107 -181
Axis I J K L M N
yi -876 -913 -950 -930 -1017.5 -1017.5
190 410 210 600 170 325
zi -351 -176 -344 -250 -250 -650
Modeling of the quarter suspension system for terrain vehicle is performed in Working
Model environments. All simulations are making with fixing vehicle body and wheel motion
in boundary position, bound and rebound is ensured by using a hydraulic actuator. Simulation
results are presented by diagrams.
Figure 4.20 presents path of wheel for both double wishbone with longer equal length
and final design of the suspension systems.
300
200
BOUND

Final design
ZM
100
REST
0 YM
-100
REBOUND
w/2
-200 LATERAL MOTION
-300
-1050 -1040 -1030 -1020 -1010 -1000 -990
longer equal control arms and final design
Figure 4.21 introduces the changing of the camber angle for both double wishbones
with longer equal length, and final design of the suspension system as a function of the
vertical wheel motion (bound and rebound). In the following chapters present, in great detail,
the final design of the terrain vehicle’s suspension system which successfully fulfilled the
design requirements given by the check list.
- 52 -
300
Final design

200
BOUND
100
REST
-100 w/2
REBOUND
-200
-300
-0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5
suspension system with double wishbone longer equal control arms and final design
Figure 4.22 show a model of the final design of the terrain vehicle suspension system
performed within the Pro/ENGINEER environment (3D view, frontal view and plane view).
a)
b)
- 53 -
c)
Figure 4.22: Final design of the terrain vehicle suspension system; [6]
a) 3D view, b) front view, and c) plane view
4.4 Discussing of the results
From Figure 4.20, it can be observed that when the wheel moves from its rest position
to a rebound position (0 ... – 250 mm), the final design of the suspension system provides 45
% less lateral displacement of the wheel. When the wheel moves from the rest to the bound
position (0 ... +250 mm) the obtained results display better behavior by ensuring 72.6 % less
displacements compared with the suspension system with double wishbone longer equal
control arms.
Therefore, the final design provides relatively small values for lateral displacements
thus effectively improving the life-expectancy of the tire, reducing the additional load on the
suspension mechanism and improving the stability of the vehicle when driving over various
terrains.
Figure 4.21 show that the change of camber angle in the final design of the suspension
system provides zero camber angles, thus causing better quality contact of the tire with the
ground surface. In this case, the tire acts perpendicularly with the ground surface. Improving
the quality contact has had an influence on improving the stability of the vehicle’s movement
when cornering, accelerating or braking.
Due to experiences, the vehicle suspension system mostly operates from rest to the
bound direction as a result of ground excitation. The final design regarding the boundary
position provides lateral wheel displacements of up to 17 mm. Such displacements in other
systems are considered to be relatively larger.
- 54 -
5 MODELING, SIMULATION AND OPTIMIZATION OF THE

SUSPENSION MECHANISM FOR A TERRAIN VEHICLE
Many centuries ago mechanisms found great application in various fields of mechanical
engineering especially when required to transfer one type of motion into another (e.g. from
displacements to rotational motion). A suitable methodology for analyzing mechanism
behavior during the development process is more than necessary in order to obtain
functionality and predefined motion. In this contest, the suspension mechanism is a spatial
mechanism and would be difficult to analyze without an effective computer program. Their
analyses are going to be more complicated when involving a greater number of rubber bushes
which have influences on the suspension mechanism links by giving variable length. Despite
these complications it is possible to gain appreciation of the suspension mechanism by
neglecting rubber bush and concentrating on their basic motion.
The objective of this chapter is the modeling, simulation, and optimization of the
suspension mechanism for a terrain vehicle. Based on the specific requirement given in the
check list, the suspension mechanism will ensure that during all vertical motion of the wheels
(±250 mm bound and rebound), the camber angel should be zero. Except camber angle it is
necessary that the lateral motion of the wheels is as small as possible when vehicle wheels
moves upward or downward, in order to improve vehicle stability.
In order to estimate these requirements, it is necessary to have a powerful tool to
compute those motions and forces that act in suspension mechanism as a result of the ground
excitation. To speed this procedure it is helpful to have an effective program that simulates
the suspension mechanism’s behavior by employing appropriate essential parameters. One of
the well-established ways of explaining mechanism performance is the modeling of the
quarter of it. The developing of computer programs for solving equations of motion is
performed in MathCAD environment for a quarter of it. Solutions are mainly based on the
kinematic and dynamic behaviors which determine the motions and loads that affect the
suspension mechanism, Werff [55]. To verify the reliability of the developed model, a
comparison with one of the simple commercial Working Model Software [56], is carried out.
In order to have an as small as possible lateral motion of the wheels, an optimization
procedure is done for getting optimal lengths of suspension mechanism links.
- 55 -
5.1 Kinematic modeling and analysis of the suspension mechanism
The kinematic analyses for the terrain vehicle suspension mechanism are estimated according
geometric modeling data shown in the final design for the suspension mechanism (chapter 4).
By these data are estimating the mechanism’s behavior. For kinematic analysis it is necessary
to check the geometric variations of the components of the suspension mechanism over the
operating range, spring and shock absorber relative motion along the vertical motion of the
wheels. Notwithstanding the apparent complexities of some types of suspension systems, a
basic understanding of their kinematic behavior can be derived from a two – dimensional
analysis, by considering the motion of it only within a vertical plane (y, z – axis), Happian-
Smith [57].
Determining the kinematic expressions for the suspension mechanism will enable
simulation of the vertical displacement, velocity, and acceleration of the mechanism links for
different positions of the wheel, for which expressions are usually inconvenient. Solving of
the kinematic relations is performed in MathCAD and MATLAB environments.
5.1.1 Displacement analysis of the suspension mechanism
Terrain vehicle suspension mechanism for displacement analysis is shown in Figure 5.1.
0
F
A
0
8
1
B
G C
0 0
z
BOUND 4 3
ω3 2 O
δw 5 E -y
zmax H D
J
6
M L 7
zmin K I
REBOUND N
Figure 5.1: Kinematic linkages of the terrain vehicle suspension mechanism
- 56 -
The mechanism presented in Figure 5.1, has nine links, such as: 0, 1, 2, 4, 5, 6, 7 and 8
links. The motions of each link is in the vertical plane (y, z – axis). To derive expressions for
computing displacement of the suspension mechanism it is necessary to employ methodology
for the projection of the vector position defined by the vector loops. Each link of the
mechanism is represented by a vector position. In followers are treated as vector loops and the
equations of motion for the suspension mechanism is shown in Figure 5.1.
5.1.1.1 Vector loops and vector position
Based on the design task, the suspension mechanism should allow the wheel to move in a
downward direction from the rest position (rebound) for displacement zmin = - 250 mm where
input angular displacement of links EG is θ3min = 335.18 deg. When the wheel moves upwards
vertically from the rest position (bound), the maximal displacement is zmax = 250 mm and
is θ3max = 378.22 deg. So, input angular displacement θ3 of the link
input angles of link
is when considering primary variable by range from 335.18 to 378.22 deg. In order to

determine the vertical motion of the wheel (point M) is necessary for determining unknown
angular displacement such as θ1(θ3), θ2(θ3), θ9(θ3) and relative length of spring and shock
absorber Ls(θ3) as a function of input angular displacement θ3. The vector positions of the
suspension mechanism are: R1, R2, R3, R4, R5, R6, R7, R8, and R9. The direction of each vector
is arbitrary, however the angle is associated with the vectors direction measured using the
positive direction of the y – axis. The unknown angular displacement and vectors position are
shown in Figure 5.2, Figure 5.5 and Figure 5.8.
z
A 0 θ1
1 y
G θ3max R4 A
C
0 θ2 R1
G
4 θ4
θ3min θ3 C
5 E E R2
R3
Figure 5.2: Vector loops of four-bar links derived from the suspension mechanism
A closed vector loop (four-bar links) is formed from the vector loops: R1, R2, R3, and
R4, Figure 5.2. Link
is formed by the vector position R4 called the ground link which is
is the vector position R3
fixed in the vehicle’s body at an angular displacementθ4. Link
which is considered as the input link and is controlled by the input angular displacement θ3.
forms vector position R2 as a coupler link with angular displacement θ2 and connects
Link
- 57 -
introduces vector position R1 which is the output link with

the input and output links. Link
angular displacement θ1. The output angular displacement θ1 and coupler angular
displacement θ2 are functions of the links length and input angular displacement θ3. The
vector loops of the four-bar links and data for links are shown in Figure 5.2 and Table 5.1.
Table 5.1 Vector representation of the four-bar linkages shown in Figure 5.2
Links Name Vector Length [mm] Angular angle [deg] Variable

Output R1 r1 = 220 θ1 → YES

Coupler R2 r2 = 190 θ2 → YES

Input R3 r3 = 210 θ3 ~335...378 → YES

Ground R4 r4 = 483.04 θ4 → NO
Each vectors position Ri is described by length ri and angular displacement θi. The
,
vector loop for four bar linkages (links: and
, ) in global coordinate is:
R1 + R2 − R3 + R4 = 0 , (5.1)
where,
R1÷ 4 = R x 1÷ 4 i + R y 1÷ 4 j = r1÷ 4 ⋅ (cos θ 1÷ 4 i + cos θ 1÷ 4 j ) . (5.2)
The horizontal and vertical components for each vector loop’s equation is required the
sum to be identically zero at all times in order to satisfy the physical assembly of the links.
Substituting the Cartesian expression for the planar vector in equation (5.1), the result in:
r1 ⋅ cosθ1 + r2 ⋅ cosθ 2 − r3 ⋅ cosθ3 + r4 ⋅ cosθ 4 = 0 ,

(5.3)
r1 ⋅ sinθ1 + r2 ⋅ sinθ2 − r3 ⋅ sinθ3 + r4 ⋅ sinθ4 = 0 ,
after regulation of expression (5.3), we get expressions:
r2 ⋅ cosθ2 = r3 ⋅ cosθ3 − r1 ⋅ cosθ1 − r4 ⋅ cosθ4 ,

(5.4)
r2 ⋅ sinθ2 = r3 ⋅ sinθ3 − r1 ⋅ sinθ1 − r4 ⋅ sinθ4 .
To derive the relationship between the input θ3 and output angular displacement θ1, it is
necessary to eliminate the coupler angular displacement θ2 by some mathematical operation
between equation (5.3) and (5.4). By squaring both sides, providing the following equations:
(r2 ⋅ cosθ2 )2 = (r3 ⋅ cosθ3 − r1 ⋅ cosθ1 − r4 ⋅ cosθ4 )2 , (5.5)
(r2 ⋅ sinθ2 )2 = (r3 ⋅ sinθ3 − r1 ⋅ sinθ1 − r4 ⋅ sinθ4 )2 . (5.6)
After some mathematic operation, expressions (5.5) and (5.6) are written as follow:
(r2 ⋅ cosθ 2 )2 = r12 ⋅ cos2 θ1 + r32 ⋅ cos2 θ 3 + r4 2 ⋅ cos2 θ 4 + K
(5.7)
+ 2 ⋅ [r1 ⋅ cos θ1 (r4 ⋅ cos θ 4 − r3 ⋅ cos θ 3 ) − r3 ⋅ r4 ⋅ cos θ 3 ⋅ cos θ 4 ],
- 58 -
(r2 ⋅ sin θ 2 )2 = r12 ⋅ sin 2 θ1 + r32 ⋅ sin 2 θ 3 + r4 2 ⋅ sin 2 θ 4 + K

(5.8)
+ 2 ⋅ [r1 ⋅ sin θ1 (r4 ⋅ sin θ 4 − r3 ⋅ sin θ 3 ) − r3 ⋅ r4 ⋅ sin θ 3 ⋅ sin θ 4 ] .
By adding equation (5.7) and (5.8), generates the following expression:
cos θ1 ⋅ ( r4 ⋅ cos θ 4 − r3 ⋅ cos θ 3 ) + sin θ1 ⋅ ( r4 ⋅ sin θ 4 − r3 ⋅ sin θ 3 ) =

2 2
r − r1 − r3 − r4
2 2
r ⋅ r ⋅ (cos θ 3 ⋅ cos θ 4 + sin θ 3 ⋅ sin θ 4 ) (5.9)
= 2 + 3 4 .
2 ⋅ r1 r1
By substitution of the trigonometric function cos(α-β), equation (5.9) is written as follow:
2 2 2 2
r2 − r1 − r3 − r4 r ⋅r
r4 ⋅ cos(θ1 − θ 4 ) − r3 ⋅ cos(θ1 − θ 3 ) = + 3 4 ⋅ cos(θ 3 − θ 4 ) . (5.10)
2 ⋅ r1 r1
Equation (5.14) is called as Freudenstein's Equation. From equation (5.10), it is difficult
to determine the value of output θ1 for the given value of input angular displacement θ3
therefore it is necessary to simplify this equation by using trigonometrically expressions:
2 ⋅ tan(θ1 / 2) 1 − tan2 (θ1 / 2)

sinθ1 = ; cosθ1 = . (5.11)
1 + tan2 (θ1 / 2) 1 + tan2 (θ1 / 2)
Substituting sinθ1, cosθ1, and some other mathematical operation in order to simplify
the relation between the inputs angles θ3 and the output angles θ1, obtains the more practical
expression given below:
θ  θ 
A(θ3 ) ⋅ tan2  1  + B(θ3 ) ⋅ tan 1  + C(θ3 ) = 0 , (5.12)
2 2
where substitution constants A(θ3), B(θ3) and C(θ3) are in a function of the input variable
angular displacement θ3, written by the following expressions:
2 2 2 2
r2 − r1 − r3 − r4 r ⋅r
A(θ 3 ) = r4 ⋅ cosθ 4 − r3 ⋅ cosθ 3 + + 3 4 ⋅ cos(θ 3 − θ 4 ) , (5.13)
2 ⋅ r1 r1
B (θ 3 ) = 2 ⋅ (r3 ⋅ sin θ 3 − r4 ⋅ sin θ 4 ) , (5.14)
2 2 2 2
r2 − r1 − r3 − r4 r ⋅r
C (θ 3 ) = + 3 4 ⋅ cos(θ 3 − θ 4 ) − (r4 ⋅ cosθ 4 − r3 ⋅ cosθ 3 ) . (5.15)
2 ⋅ r1 r1
As shown equation (5.12) is a quadratic function in tan(θ1/2) and their solution for
finding the output angular displacement θ1 are written as follow:
2
 θ1  − B(θ 3 ) ± B (θ 3 ) − 4 ⋅ A(θ 3 ) ⋅ C (θ 3 )
tan  = . (5.16)
2 2 ⋅ A(θ 3 )
Finally output angular displacement θ1, may be calculated by expression:
- 59 -
 − B (θ 3 ) ± B 2 (θ 3 ) − 4 ⋅ A(θ 3 ) ⋅ C (θ 3 ) 
θ1 (θ 3 ) = 2 ⋅ tan −1  . (5.17)
 2 ⋅ A(θ 3 ) 
rotates for small angles to satisfy the
The input angular displacement θ3 of the link
vertical motion of the wheel. The data for determining the output angular displacement θ1 are
shown in Table 4.5 & Table 5.1. Obtained results are performed in the MathCAD [58]
environment and exported in Microsoft Excel graphic presentation, Figure 5.3.
272
A
268 θ3rest =
355.40 [deg] θ1 =260.60 [deg]
Output angular displacement θ1 [deg]
264 E
G C
260
256
A A
θ1 =
252 θ3max = 254.71 [deg]
378.22 [deg]
248 G θ1 =
249.09 [deg] E
C
C G
244 θ3min =
335.18 [deg] E
240
335 339 343 347 351 355 359 363 367 371 375
Input angular displacement θ3 [deg]
Figure 5.3: Output angular displacement θ1 as a function of input angular displacement θ3
To find the coupler angular displacement θ2 it is necessary to eliminate output angular

displacement θ1 from equation (5.3). So, by squaring both sides and simplifying the
expression by using trigonometric transformation, equation (5.3) now is:
2 2 2 2
r −r −r −r r ⋅r
r4 ⋅ cos(θ2 − θ4 ) − r3 ⋅ cos(θ2 − θ3 ) = 1 2 3 4 + 3 4 ⋅ cos(θ3 − θ4 ) . (5.18)
2 ⋅ r2 r2
Similar to previously, equation (5.18) is unsuitable for obtaining the coupler angular
displacement θ2 and it is necessary to transform in a more favorable form such as:
θ  θ 
D(θ3 ) ⋅ tan2  2  + E(θ3 ) ⋅ tan 2  + F (θ3 ) = 0 , (5.19)
2 2
- 60 -
where the substitution constants D(θ3), E(θ3) and F(θ3) are as a function of the input variable
angular displacement θ3, determined by the following expressions:
2 2 2 2
r − r2 − r3 − r4 r ⋅r
D(θ 3 ) = r4 ⋅ cosθ 4 − r3 ⋅ cosθ 3 + 1 + 3 4 ⋅ cos(θ 3 − θ 4 ) , (5.20)
2 ⋅ r2 r2
E (θ 3 ) = 2 ⋅ (r3 ⋅ sin θ 3 − r4 ⋅ sin θ 4 ) , (5.21)
2 2 2 2
r − r2 − r3 − r4 r ⋅r
F (θ 3 ) = 1 + 3 4 ⋅ cos(θ 3 − θ 4 ) − (r4 ⋅ cos θ 4 − r3 ⋅ cos θ 3 ). (5.22)
2 ⋅ r2 r2
Equation (5.19) is a quadratic function in tan(θ2/2) and is used to find the coupler
angular displacement θ2, through the following expressions:
 − E(θ3 ) ± E 2 (θ3 ) − 4 ⋅ D(θ3 ) ⋅ F (θ3 ) 

θ2 (θ3 ) = 2 ⋅ tan−1  . (5.23)
 2 ⋅ D(θ3 ) 
By using the equations (5.17) and (5.23), the output θ1 and coupler θ2 are defined as a
function of the input angular displacement θ3, by data given in Table 5.1.
220
A
212
θ3rest =355.40 [deg]
Coupler angular displacement θ2 [deg]
204
E
196
G C
θ2 =182.96 [deg]
188
180
A A
172
θ3max =
164 378.22 [deg]
G C E
156 θ2 =
E
G 159.08 [deg] C
θ3min = θ2 =
148 249.17 [deg]
335.18 [deg]
140
335 339 343 347 351 355 359 363 367 371 375
Figure 5.4: Coupler angular displacement θ2 as a function of input angular displacement θ3
The piston rod of the shock absorber known as the slider mechanism is fixed in the
vehicle’s body by a revolute joint F. The bottom of the spring & shock absorber is connected
- 61 -
which is connected to the upper control arm of

by revolute joint E and single arm - link
is also fixed in the vehicle’s body by a revolute joint
the suspension mechanism. The link
G. The vector loops of this configuration are shown in Figure 5.5.
F θ9
F
R8
8 R9
G θ3max G θ8
θ3
4
θ3min E R3 E
Figure 5.5: Vector loops of three-bar links derived from the suspension mechanism
Vector loops are formed from vectors position such as loops: R3, R8 and R9. The
direction of each vector is arbitrary. The vector position - link
is R8 known as ground
fixed link in a vehicle body defined by angular displacement θ8. The other position vector -
is R3 considered as an input link for the suspension mechanism which is controlled by
link
is the slider link (stroke of the sprig & shock
the input angular displacement θ3. Link
absorber) which produced variables vector position R9 with angular displacement θ9. The
is considered as output link Ls which connects both the input and output links.
slider link
The slider link Ls and angular displacement θ9 are in the function of the input variables
angular displacement θ3. For the vector position that connects ground link r8 and input link r3
will remain constant value. The slider link Ls and angular displacement θ9 will change as the
results of the wheel motion caused from the suspension mechanism motion. The vector loop
is written as follows:
R8 + R9 − R3 = 0 , (5.24)
where,
R3,8 , 9 = R x 3,8 , 9 i + R y 1÷ 4 j = r3,8, 9 ⋅ (cos θ 3,8, 9 i + cos θ 3,8 , 9 j ) . (5.25)
By decomposition the vector loop equation (5.24) into sinθi and cosθi, obtains this expression:
r8 ⋅ cos θ 8 + Ls ⋅ cos θ 9 − r3 ⋅ cos θ 3 = 0 , (5.26)
r8 ⋅ sin θ 8 + L s ⋅ sin θ 9 − r3 ⋅ sin θ 3 = 0 . (5.27)
To derive at the relation between the input θ3, output length Ls and angular
displacement θ9, it is necessary to eliminate the output angular displacement θ9 by some
mathematical rearrangements between equations (5.26) and (5.27), written as follows:
- 62 -
2 2 2 2
Ls (θ3 ) = r3 + r8 − 2 ⋅ r3 ⋅ r8 ⋅ cos(θ3 − θ9 ) . (5.28)
Figure 5.6 presents the slider length s of the spring & shock absorber in the function of
the input angular displacement θ3 determined from equation (5.28). The vector representation
for the three bar links with necessary data for suspension mechanism is shown in Table 5.2.
Table 5.2 Vector representation of the three-bar linkages shown in Figure 5.5
Links Name Vector Length Angular displacements Variable

Input R3 r3 = 210 θ3 → YES

Ground R8 r8 = 423.03 θ8 = 68.65 → NO

Output R9 Ls θ9 → YES
520
F
500
Length of spring & shock absorber Ls [mm]
Ls = 414.56
480
460
G
440
REST POSITION
E
420 θ3rest = 355.40 [deg]
F BOUND F
400
Ls = 331.51
Ls = 483.53 [mm]
380
REBOUND
360
340 G G E
θ3max =
320 θ3min = 378.22 [deg]
335.18 [deg] E
300
335 339 343 347 351 355 359 363 367 371 375
Input angular displacementθ3 [deg]
Figure 5.6: Length of spring-shock absorber Ls as a function of input angular displacement θ3
The relation between output angular displacement θ9 and input angular displacement θ3
is determined after some mathematical arrangement into equations (5.26), (5.27) and (5.28).
The output angular displacement θ9 is defined by the following expression:
r3 ⋅ cosθ 3 − r8 ⋅ cosθ 8
θ 9 (θ 3 ) = . (5.29)
2 2 2 2
r3 + r8 − 2 ⋅ r3 ⋅ r8 ⋅ cos(θ 3 − θ 9 )
The equations (5.28) and (5.29) determine the variable output length Ls (Figure 5.6) and
angular displacement θ9 of the slider link as functions of the input angular displacement θ3.
- 63 -
Figure 5.7 presents the output θ9 as a function of the input angular displacement θ3.
291
F
θ9 =277.67 [deg]
288
Output angular displacement θ9 [deg]
285
282
θ3rest =355.40 [deg]
G
E REST POSITION
279
276
θ9 =277.89 [deg]
F F
θ9 =274.34 [deg]
273
BOUND
270
267 G E
G REBOUND
θ3max =
264
θ3min = 378.22 [deg]
335.18 [deg] E
261
335 339 343 347 351 355 359 363 367 371 375
Figure 5.7: Output θ9 as a function of input angular displacement θ3
Figure 5.8 shows closed-loops for four-bar linkages. It is formed by vectors position:
is formed by vector position R7 called a ground link fixed in the
R1, R5, R6 and R7. Link
is described with vector position R1

vehicle’s body at angular displacement θ7. Link
considered as input link respectively output link (Figure 5.2) and is controlled from input
determines vector position R5 known as coupler link with
angular displacement θ3. Link
introduces vector
angular displacement θ5 that connect both input and output links. Link

position R6 considered as the output link with angular displacement θ6.
A z θ1
θ1 0 θ7
A y
1 R 7 θ6
B R1
0
C θ5
3 C B
2 R5 R6
D D
Figure 5.8: Vector loops of four-bar links derived from the suspension mechanism
- 64 -
The vector representation for the four-bar links with necessary data for suspension
mechanism is shown in Table 5.3.
Table 5.3 Vector representation of the four-bar linkages shown in Figure 5.8
Links Name Vector Length [mm] Angular displacement [deg] Variable

Input (Output) R1 r1 = 220 θ1 → YES

Coupler R5 r5 = 170 θ5 = 270 → NO

Output R6 r6 = 220 θ6 = θ1 → YES

Ground R7 r7 = 170 θ7 = 270 → NO
Table 5.3 shown that the lengths of links: r1+r5 are equal with r6+r7 then by using Grashof’s
criterion of movability, the four bar mechanism can be classified into a parallelogram
mechanism as a double crank mechanism. This means that by changing the input θ1 of link
also output angular displacement θ6 of link

has same angles θ1=θ6. The angular
don’t change during its motion because link
displacement θ5 of link and ground link
are equal and parallel and has the same values as ground angular displacement θ7.

By using such a mechanism as an additional part of the suspension mechanism makes it

possible that the orientation of the coupler angular displacement does not change during all
the motion. Maintaining the parallelism is an important thing due to the fact that the vehicle
suspension mechanism must maintain the chamber angle at zero.
5.1.1.2 Determining of the wheel path and other characteristic of the suspension mechanism
By determining the angular displacement and lengths of each link defined previously is
possibility of determining the positions of each link in the y and z axes of the suspension
mechanism, respectively the positions of each joint. Specific interest in this approach is
determining the positions of the center M and the bottom N of the wheel, defined as follows:
Y E (θ 3 ) = y A + r1 ⋅ cos(θ 1 ) + r2 ⋅ cos(θ 2 ),
(5.30)
Z E (θ 3 ) = z A + r1 ⋅ sin(θ1 ) + r2 ⋅ sin(θ 2 ),
Y M (θ 3 ) = y A + r1 ⋅ cos(θ 1 ) + r2 ⋅ cos(θ 2 ) + EH ⋅ cos(θ 2 ) + ( y M − y H ) ,

(5.30)
Z M (θ 3 ) = z A + r1 ⋅ sin (θ 1 ) + r2 ⋅ sin (θ 2 ) ,+ EH ⋅ sin(θ 2 ) + ( z M − z H ) ,
YN (θ 3 ) = y B + r6 ⋅ cos(θ 6 ) + DI ⋅ cos(θ10 ) + ( y N − y I ) ,
(5.31)
Z N (θ 3 ) = y B (θ 3 ) + r6 ⋅ sin(θ 6 ) + DI ⋅ sin(θ10 ) + ( z N − z I ) .
Expression (5.30) determines the paths or motions at the center of wheel (YM(θ3) and
ZM(θ3)) as a function of the input angular displacement θ3. The essential data for determining
the paths of the center of wheel (YM(θ3), ZM(θ3) and ZM = f(YM) ) are given in Table 4.5, Table
5.1 and Table 5.3. Figures 5.9 to 5.11 presents motion of the center of wheel.
- 65 -
300
Vertical motion of center of wheel ZM(θ3) [mm]

BOUND
200
100
BOUND
0 REST POSITION θ3rest
zmax ZM G
H
E
-100 REST POSITION
M
zmin I
-200
REBOUND
REBOUND
-300
330 335 340 345 350 355 360 365 370 375 380
0
Input angular displacement θ3 [ ]
Figure 5.9: Vertical motion of center of wheel as a function of input angular displacement θ3
-1200
Lateral motion of center of wheel YM(θ3)
-1100 REST POSITION
-1000
REBOUND BOUND BOUND
θ3rest
-900 ZM G
zmax H
E
[mm]
-800 REST POSITION M

YM
zmin I
-700
REBOUND
-600
-500 YM
-400
330 335 340 345 350 355 360 365 370 375 380
0
Input angular displacement θ3 [ ]
Figure 5.10: Lateral motion of center of wheel as a function of input angular displacement θ3
The expression for the determining kingpin inclination angle (σ) of the suspension
mechanism, is written as follow:
 ( y K − y I ) − ( y J − y H )
σ = a tan   , [deg] (5.32)
 (z I − z H ) 
By substitution of known values, kingpin inclination angle is: σ = 12.28 [deg].
- 66 -
300
Vertical motion of center of wheel ZM(θ3)

200
BOUND
100 θ3rest
zmax ZM G
[mm] H Z
E
0 REST POSITION M
YM O
zmi
-Y
I
-100
REBOUND
-200
YM
-300
-1100 -1000 -900 -800 -700 -600 -500 -400
Lateral motion of wheel YM(θ3) [mm]
Figure 5.11: Path of the center of wheel of the suspension mechanism
The expression for determining the kingpin offset is written as follows:
Kingpin offset = [ y M − y K + ( z M − z K − rd ) ⋅ tan σ ], [mm] (5.33)

By substitution of known values, kingpin offset is: Kingpin offset = 0.9 [mm].
Whilst the expression for the determining of camber angle γ is derived from expressions
(5.30) and (5.31) as a function of the input angular displacement θ3, is written as follow:
 YM (θ 3 ) − YN (θ 3 ) 
γ (θ 3 ) = a tan   , [deg] (5.34)
 Z M (θ 3 ) − Z N (θ 3 ) 
By the substitution of known values, the camber angle during all vertical motion of the
wheel is zero: γ (θ3) = 0 [deg].
5.1.1.3 Comparison of the obtained results between developed program and the commercial
Working Model Software
To be sure when developing the mathematical model performed in MathCAD
environments, in Working Model software [53] the numerical simulations are performed for
the same known geometric data of the suspension mechanism. Simulation results are shown
for wheel path and other significant characteristic of the suspension mechanism. Table 5.4
presents the results obtained by two ways in order to compare the accuracy of the developed
model. Result are presented when the wheels of the terrain vehicle are in rebound
ZMmin = - 250 mm, rest ZMrest = 0 and bound position ZMmax = 250 mm.
- 67 -
Table 5.4: Comparison of the obtained result between the developed programs
performed in MathCAD and the simply commercial Working Model Software
Angular displacement and Wheel motion and displacement of
Input angle camber angle [deg] the joint E and slider length s [mm]
[deg]
θ1(θ3) θ2(θ3) θ9(θ3) γ(θ3) YM(θ3) ZM(θ3) YE(θ3) ZE(θ3) s(θ3)
θ3min = - - -
335.18 249.09 209.18 274.34 0.00 1017.50 250.05 517.40 18.85 483.54
- -
Theory
θ3rest =
355.40 260.60 182.96 277.67 0.00 1050.00 0.00 498.68 90.153 414.56
θ3max = - -
378.22 254.71 159.08 277.89 0.00 1033.00 250.05 508.52 172.64 331.49
θ3min = - - -
Working Model
335.18 249.11 209.14 274.34 0.00 1017.50 250.00 517.40 18.93 483.45
θ3rest = - - -
355.40 260.60 182.95 277.67 0.00 1050.14 0.21 498.68 90.17 414.54
θ3max = - -
378.22 254.70 159.08 277.89 0.00 1033.51 249.72 508.52 172.62 331.52
Figure 5.12 shown the obtained results within the Working Model environment when the
wheel of the terrain vehicle is in the rest position. The same results are tabulated in Table 5.4.
Figure 5.12: Modeling of the suspension mechanism in WM environment
From Table 5.4, it can be observed that the obtained results from the developed program
in the MathCAD environment coincides very well with the results performed by commercial
Working Model software. Consequently, it can be concluded that, the developed program is
suitable for further analysis of the terrain vehicle suspension mechanism.
- 68 -
5.1.2 Velocities analysis of the suspension mechanism
Assuming that the input variable angular displacement θ3 and angular velocity ω3(t) are
predefined and angles and length θ1(θ3), θ2(θ3), θ9(θ3) and Ls(θ3) are determined by equations
(5.17), (5.23), (5.29), and (5.28). Therefore, the angular velocities ω1(t), ω2(t) and ω3(t) may
be found by differentiating equations (5.3), (5.26) and (5.27) with respect to time and after
some mathematical operation, can be obtain by the following expressions:
r3 sin(θ 2 − θ 3 )
ω1 (t ) = ⋅ ⋅ ω3 ( t ) , (5.35)
r1 sin(θ 2 − θ1 )
r3 sin(θ1 − θ 3 )
ω2 (t ) = ⋅ ⋅ ω3 (t ) , (5.36)
r2 sin(θ1 − θ 2 )
r3 ⋅ cos(θ 3 − θ 9 )
ω9 (t ) = 2 2 2 2
⋅ ω3 (t ) , (5.37)
r3 + r8 − 2 ⋅ r3 ⋅ r8 ⋅ cos(θ 3 − θ 9 )
v Ls (t ) = r3 ⋅ sin(θ 9 − θ 3 ) ⋅ ω3 (t ) , (5.38)
when are:
dθ1 & dθ
ω1 (t ) = = θ1 ; ω2 (t ) = 2 = θ&2 - Angular velocity of the link
and

dt dt
dθ dθ
ω3 (t) = 3 = θ&3 ; ω9 (t) = 9 = θ&9 - Angular velocity of the link and

dt dt
dL •
vLs (t ) = s = Ls - Relative velocity of the shock absorber-slider link
dt
By determining the values derived from equations (5.35) to (5.38), results can be
obtained for the velocities of each link and joint of the suspension mechanism. Particular
interest is shown in finding the velocity at the center of wheel vM(t), determined by the
following expression:
vM (t ) = [vMY (t)] 2 + [vMZ (t )] 2 , (5.39)

when vMY(t) denotes the component of velocity in y, whilst vMY(t) is component of velocity in z
- axis, determined by deriving equation (5.30) with respect to time, given as follow:
dYM (θ3 )
vMY (t ) =
dt
[ ( ) ]
= − r1 ⋅ sin(θ1 ) ⋅ ω1 (t ) + r2 + EH ⋅ cos(θ2 ) ⋅ ω2 (t ) , (5.40)
dZ (θ )
( )
vMZ (t ) = M 3 = r1 ⋅ cos(θ1 ) ⋅ ω1 (t ) + r2 + EH ⋅ sin(θ2 ) ⋅ ω2 (t ) .
dt
(5.41)
Equation (5.39) determines absolute velocity of the center wheel vM(t). This velocity
will be obtained when the vehicle body is fixed and the wheel moves relatively around its
- 69 -
positions throughout the suspension mechanism. The situation is completely different when
the vehicle’s body and wheels moves at the same time, caused by ground excitation. This case
is treated in the next chapter.
5.1.3 Accelerations analysis of the suspension mechanism
Assuming that, the angular displacement θ3, angular velocity ω3(t) and angular acceleration
. Then the angles and lengths θ1(θ3),
ε3(t) are predefined as the kinematic of the input link
θ2(θ3), θ9(θ3) and Ls(θ3) are determined by equations (5.17), (5.23), (5.29), and (5.28), as well
as ω1(t), ω2(t), ω9(t) and vLs(t) being determined by equations (5.35) to (5.38). Therefore, the
angular acceleration ε1(t), ε2(t), ε9(t) respectively linear acceleration aLs(t), may be found by a
second differentiation of the equations (5.3), (5.26) and (5.27) with respect to time and after
some mathematical regulation can be obtained by the following expressions:
2 2 2
r ⋅ ω ⋅ cos(θ1 − θ 2 ) + r2 ⋅ ω2 − r3 ⋅ ω3 ⋅ cos(θ 2 − θ 3 ) r3 ⋅ sin(θ3 − θ 2 )
ε 1 (t ) = 1 1 − ⋅ ε 3 (t ) , (5.42)
r1 ⋅ sin(θ 2 − θ1 ) r1 ⋅ sin(θ 2 − θ1 )
2 2 2
r1 ⋅ ω1 + r2 ⋅ ω2 ⋅ cos(θ1 − θ 2 ) − r3 ⋅ ω3 ⋅ cos(θ1 − θ 3 ) r3 ⋅ sin(θ 3 − θ1 )
ε 2 (t ) = − ⋅ ε 3 (t ) , (5.43)
r2 ⋅ sin(θ1 − θ 2 ) r2 ⋅ sin(θ1 − θ 2 )
2
r ⋅ sin(θ3 + θ9 ) ⋅ ω3 − 2 ⋅ vs ⋅ ω9 r3 ⋅ cos(θ3 + θ9 )
ε 9 (t ) = 3 − ⋅ ε 3 (t ) , (5.44)
s s
2
aLs (t) = L&&s (t) = Ls ⋅ ω9 − r3 ⋅ ω3 ⋅ cos(θ3 −θ9 ) − r3 ⋅ sin(θ3 −θ9 ) ⋅ ε3 (t) . (5.45)
when are:
dω1 && dω
ε1(t) = = θ1; ε 2 (t) = 2 = θ&&2 - Angular acceleration of the link
and

dt dt
dω dω
ε 3 (t ) = 3 = θ&&3; ε 9 (t ) = 9 = θ&&9 - Angular acceleration of the link
and

dt dt
dvLs d 2 Ls &&
a Ls (t ) = = 2 = Ls (t ) - Relative acceleration of the spring & shock absorber
dt dt
By the simulation of the equations (5.42) to (5.45) the results of acceleration can be
found for each links and joints of the suspension mechanism. Let us now try to find an
equation for determining the acceleration at the center of wheel aM(t), determined by the
aM (t ) = [aMY (t )] 2+ [aMZ (t)] 2 (5.46)

when aMY(t) denotes the component of acceleration in y - axis, whilst aMY(t) is the component
of acceleration in z - axis, performed by deriving equations (5.40) and (5.41) with respect to
time, generate by the following expressions:
- 70 -
dvMY (t )
aMY (t ) = ( )[ 2
= r2 + EH ⋅ sin(θ2 ) ⋅ ω2 (t ) − cos(θ 2 ) ⋅ ε 2 (t ) − K]
dt (5.47)
[ 2
− r1 ⋅ cos(θ1 ) ⋅ ω1 (t ) + sin(θ1 ) ⋅ ε1 (t ) ,]
dvMZ (t )
a MZ (t ) = ( )[ 2
]
= r2 + EH ⋅ cos(θ 2 ) ⋅ ω2 (t ) + sin(θ 2 ) ⋅ ε 2 (t ) − K
dt (5.48)
[ 2
− r1 ⋅ sin(θ1 ) ⋅ ω1 (t ) − cos(θ1 ) ⋅ ε1 (t ) .]
Equation (5.46) determines the absolute acceleration at the center of wheel aM(t). This
case is valid when the vehicle’s body is fixed and the wheel moves relatively around its
positions.
5.2 Dynamic modeling of forces that acts on the suspension mechanism
There are two fundamental forms of forces that act on the suspension mechanism of the
vehicle. The first type of force is internal force such as the reactions forces of joints Fi,
spring forces Fs and shock absorber damper forces Fsh. The reaction forces are as a result of
the joints of two bodies (connection between two links). Due to the principle of the third
Newton’s law, these forces are often called action-reaction forces F12 = - F21, as shown in
Figure below:
F12 F21
1 2
The spring force Fs is a force that act along the spring axis (axis between joints E and
F), schematically shown in Figure 5.13.
a) θ9(t) Fs b) θ9(t) c) θ9(t)

F F F
Fs
Ls(com)
Ls(free)
Ls(ext)
Fs Fs
E
E
COMPRESSION CASE FREE LENGHT CASE EXTENSION CASE
Positive force: + Fs Zero force: Fs = 0 Negative force: - Fs
Figure 5.13: Working principles of the spring; a) compression, b) free and c) extension
- 71 -
The forces that act between joints E and F are equal but in opposite directions.
Determining the spring force is based on the assumption that the spring force is linear in
nature. The spring force is a function of the spring length LS and spring stiffness ks. Therefore,
the spring force is determined by Hooke’s law, defined by the following expression:
Fs = k s ⋅ [Ls − Ls (def )], [N] (5.49)

where LS denotes free length of the spring at zero force, whilst Ls(def) is the length of the
spring in compression or extension cases (working length) when force is applied.
From equation (5.49) it seems that the spring force will be positive when the spring is in
compression case, zero when the length of spring is free, whilst negative when it is in
extension case. A presentation of the spring forces Fs is shown in Figure 5.14. The convention
sign used for spring force is positive +Fs when the spring is in compression case, whereas
negative -Fs when in extension case. This convention sign is opposite to the convention used
for stress analysis and finite element analyses programs, Blundell and Harty [59].
Fs [N]
FREE LENGHT (L)
+ Fs + Fs
- Fs - Fs
COMPRESSION
CASE
Ls
EXTENSION
Fs = k ⋅ [ L − Ls ( def )] CASE
-Fs [N]
Figure 5.14: Convention sign used for a linear spring force as a function of spring length
The shock absorber damping force Fsh is a function of relative velocity of piston vLs and
shock absorber damping coefficient csh. Therefore, the shock absorber damping force is
determined by the following expression:
Fsh = −csh ⋅ vLs , [N] (5.50)

So, the force generated in the shock absorber is related to the relative velocity acting
along the shock absorber axis (axis between joints F and E). The shock absorber damping
force is positive +Fsh when the shock absorber is in compression case (bound motion) and
negative -Fsh when in extension case (rebound motion), graphically shown in Figure 5.15.
- 72 -
Fsh [N]
Fsh = −c sh ⋅ v Ls Fsh < 0, vLs > 0
F E
- Fsh - Fsh
COMPRESSION Piston
CASE
- vLs O + vLs
F E EXTENSION
CASE
-Fsh, [N]
Fsh Fsh
Fsh > 0, vLs < 0
Figure 5.15: Convention sign used for a linear damper force as a function of piston velocity
The shock absorber damping force is assumed to have linear characteristics whilst most
of the shock absorbers exhibit non-linear relations between force and relative velocity. The
shock absorber damping force is not only non-linear but also asymmetric, having different
characteristics in bound and rebound motions, Blundell and Harty [59]. Spring and shock
absorbers are usually mounted in parallel and will be used in the passive suspension system.
The second types of forces applied on the suspension mechanism are external forces.
External forces are: gravitational forces and other forces caused from wheel loads, generated
forces by uneven road, cornering, braking of vehicle, which act in the tire and after them
follow through the wheel centers into the vehicle suspension mechanism.
Wheel loads may be in these directions: longitudinal (acceleration and breaking forces),
lateral (cornering forces) and vertical (gravity force caused from weight of vehicle, generated
forces by uneven road). In order to determine wheel loads it is necessary to make some
assumptions such as the chassis of the vehicle being rigid, the vehicle is driven on flat roads
(no slope), constant speed cornering, constant longitudinal acceleration, Figure 5.16 shows
the wheel loads applied on the front and rear axles for a vehicle during a braking case.
C
max G ax
Z
hCG
mg
FFx FRx
X
FFz lF lR FRz
l
Figure 5.16: Free - body diagram of a vehicle during braking (longitudinal weight transfer)
- 73 -
The loads applied on the front and rear axles are determined by using Figure 5.16 and
an equation of static equilibrium:
∑X = 0 ; ⇒
i FFx + FRx = m ⋅ ax , (5.51)
∑Z = 0; ⇒
i FFz + FRz − m⋅ g = 0, (5.52)
∑M = 0; ⇒CG FFz ⋅ lF − FRz ⋅ lR − (FFx + FFR ) ⋅ hCG = 0. (5.53)
After some mathematical substitutions, the obtain expressions for determining vertical
loads applied into the front FFz and rear axles FRz are determined by the following expression:
m ⋅ g ⋅ lR m ⋅ ax ⋅ hCG
FFz = + , (5.54)
l l
m ⋅ g ⋅ lF m ⋅ ax ⋅ hCG
FRz = − . (5.55)
l l
The first term on the right sides of the equations (5.54) and (5.55) is the static load,
whilst the second term increases the loads at the front axle, respectively decrease the loads at
the rear axle, due to braking process.
The vertical’s tire forces acting on the left or right of the front and rear axles when the
vehicle is at the rest state, are derived from expression (5.54) and (5.55):
m ⋅ g ⋅ lR m ⋅ g ⋅ lF
FFW = ; FRW = . (5.56)
2⋅l 2⋅l
Determining of the longitudinal braking force FFx and FRx from equation (5.51) and
(5.53) is impossible because we have a case of indeterminacy with four unknown forces and
there are only three equations of static equilibrium. The solution is found by the relation
between the braking and vertical loads. It is assumed that the braking system has been
designed to give same braking effort in the front and rear tires given through the coefficient of
friction µ between tire and road surface, determined by the following expressions:
 m ⋅ g ⋅ l R m ⋅ a x ⋅ hCG 
FFx =  + ⋅µ
 l l 
(5.57)
 m ⋅ g ⋅ l F m ⋅ a x ⋅ hCG 
FRx =  − ⋅µ
 l l 
where FFx, FRx denotes the braking force in the front and rear tires, FFz, FRz are the vertical
forces acting in the front and rear tires, FFW FRW introduces the vertical static forces acting in
one tire in the front and rear tires, whilst ax is termed longitudinal vehicle acceleration.
Blundell and Harty [59] and Popa [60] provide some reference values for dynamic loads
that receive tires during different driving condition of the vehicle. In fact, dynamic loads are
- 74 -
very difficult to quantifying exists to standardize what quantifies these loads. Therefore, some
manufacturers implement their own procedures for quantifying dynamic loads. Reference
values for dynamic load are given in Table 5.5, Blundell and Harty [59].
Table 5.5: Dynamic load for different driving condition, Blundell and Harty [59]
Dynamic loads on a wheel of the terrain vehicle

Dynamic load case Longitudinal Lateral Vertical
Fx [N] Fy [N] Fz [N]
1g rest (static) position 0 0 2820
3g bound movement 0 0 8461
2g rebound movement 0 0 - 5641
1) 1)
0.75g cornering (outer wheel) cases 0
1) 1)
0.75g cornering (inner wheel) cases 0
1) 1)
1g braking 0
From Table 5.5 it can be observed that, the large value of the dynamic force that the tire
receives and then transmits the suspension mechanism, is when the vehicle moves through
uneven road (bump movement) and takes values up to 3g2). Therefore, the suspension
mechanism for a terrain vehicle must withstand these dynamic loads.
5.2.1 Dynamic equation of the free bodies
For each free body the dynamic equation of motion can be written independently from other
bodies, once applied forces are known. The dynamic equations of the free body are written e
corresponding to its degrees of freedom. The total applied force on the body (Σ Fi), is given
by inertial frame (miai = Σ Fi). Its components along Yi and Zi are determined as Fyi and Fzi.
So, for each free body of planar mechanism, there are three dynamic equations of motion,
defined by the following expressions:
mi ⋅ ayi = ∑Yi = ∑Fyi , (5.58)
mi ⋅ azi = ∑Zi = ∑Fzi , (5.59)

+ CG + CG
J ci ⋅ ε i = ∑ M ci = ∑ M xi . (5.60)
Whilst in matrix form, the equations of motion determined by equations (5.58) to (5.60), are:
1)
Dynamic loads are dependent from driving condition of the vehicle such as vehicle speed, turning radius, road
slope, road surface, and vehicle height from gravity centre and vehicle acceleration;
2)
g = 9.806 [m/s2] – gravitational acceleration.
- 75 -
mi 0 0  a yi   ∑ Fyi 
0  
mi 0  ⋅  azi  =  ∑ Fzi  , (5.61)
   
 0 0 J ci   ε i  ∑ M xi 
where mi is the mass of each free body, ayi, azi denote the component of acceleration along yi
and zi axes, ΣFyi,ΣFzi introduces the sums of the components of internal and external forces
along yi and zi axes, εi is the angular acceleration for i body, Jci is mass moments of inertia for
i body, and Σ Mxi is the sum of moments acting on i body of the suspension mechanism.
The expressions (5.58) to (5.61) indicates that for dynamic analyses the physical
properties of the suspension components should be known, such as, mass mi, mass moments
of inertia Jci, mass center location Ci(yi, zi), etc. In appendix A are presents free body diagram,
of forces that act in components of mechanisms as well as expressions for determining
reaction forces in all joints by employing kinematic modeling of the suspension mechanism.
5.3 Simulation of the terrain vehicle suspension mechanism
In order to determine forces that act on the components of suspension mechanism as a result
ground excitation is started by the assumption that the ground configuration will change over
time. The ground configuration is selected in such a way that the center of the wheel reaches
the limits of the vertical motion, such as rebound zmin = - 250 mm, rest zrest = 0 and bound zmax
= + 250 mm, referring to the z – axis, Figure 5.17.
A
C1
G C B
z
C3 C4 C5
E
H C2 CG
D y
zM M
C6
C7 +zr [mm]
400
I
BOUND
zr N
Ground REST
UPWARD t [s]
REBOUND
Hidraulic Road input
DOWNWARD -zr [mm]
actuator
Figure 5.17: Simulation of ground configuration by using a hydraulic actuator
- 76 -
Simulation of the ground configuration is done by installation of a hydraulic actuator at

the bottom of the tire. The tire is assumed to be rigid and moves vertically and parallel to the
z – axis of the ground reference frame as schematically shown in Figure 5.17. The hydraulic
actuator expects to perform vertical motion of the wheel to also produce adequate driving
force which ensures the defining of kinematic and dynamic characteristics of the suspension
mechanism. The vertical displacement produced by a hydraulic actuator which represents the
configuration of the ground excitation is written by the following expression:
zr (t ) = −400 − 250⋅ cos(t ) , (5.62)

where zr(t) denotes ground excitation (road), whilst t is the termed times.
Simulation of ground configuration is done for periods of time starting from t = 0 … 2π
s. For time t = 0 & 2π s, the position of the center of wheel reaches the rebound position
zmin = - 250 mm, for time t = ½ π & ¾ π s, the position of the center of wheel is in the rest
position zrest = 0 mm, whilst for time t = π s, the position of the center of the wheel reaches
the bound position: zmax = 250 mm. Figure 5.18 illustrates ground excitation (terrain
configuration) performed by the expression (5.62) and vertical motions at the center of the
wheel as a function of time.
300
ZM(t) - motion of wheel
200 zmax
zr(t) - ground excitation
Bound
100
Vertical motion of wheel [mm]
Rest Rest
0
zrest zrest
-100 Rebound Rebound
-200
zmin zmin
-300
-400 GROUND
-500 CONFIGURATION
-600
-700
0 π /4 π /2 3π / 4 π 5π / 4 3π / 2 7π / 4 2π
Time [s]
Figure 5.18: Ground excitation zr(t) and motion of wheel ZM(t) as a function of time
In order to obtain kinematic and dynamic results for the suspension mechanism
previously known data and other characteristics for it. Table 5.6 shows the necessary data for
simulation of the terrain vehicle suspension mechanism.
- 77 -
Table 5.6: Data for simulation of the suspension mechanism
Nr. Characteristics Value Unit Nr. Characteristics Value Unit

1. Length of link 220 [mm] 15. Mass of link -m4 2 [kg]
2. Length of joints
170 [mm] 16. Mass of link

-m5 4 [kg]
3. Length of link
220 [mm] 17. Mass of link
-m6 6 [kg]
4. Length of link 170 [mm] 18. Mass of knuckle -m7 35 [kg]
5. Length of link
190 [mm] 19. Mass of inertia -JC1 16670 [kgmm2]
6. Length of link 600 [mm] 20. Mass of inertia -JC2 182000 [kgmm2]
8. Length of joints 170 [mm] 22. Mass of inertia -JC4 5083 [kgmm2]
10.
Distance ′ 141.5 [mm] 24. Mass of inertia -JC6 182000 [kgmm2]
11.
Distance 400 [mm] 25. Mass of inertia -JC7 143000 [kgmm2]
12.
Mass of link -m1 4 [kg] 26. Spring stiffness -ks 141.40 [N/mm]
13. -m2
Mass of link 6 [kg] 27. Damping coeff. -csh 14.14 [Ns/mm]
14. -m3
Mass of link 4 [kg]
To determine forces that act on the suspension mechanism, it is firstly necessary to
predefine the kinematic characteristics. By using the given data in Table 5.6 the positions,
velocities and acceleration of each joint and points of the suspension mechanism will be
determined. Besides these characteristics, the angular displacement, angular velocities and
angular acceleration all of that as a function of time, can also be determined. There are large
numbers of kinematic characteristic that can be obtained through expressions developed
previously. Therefore, we can present larger numbers of diagrams. The best presentation of
the result would be through tables for wheel center positions when in rebound, rest, and
bound. The same tables give the kinematic values obtained from numerical simulation
realized in Working Model environment. Table 5.7, Table 5.8 and Table 5.9 present the
results obtained by two ways (developed program & commercial Working Model Software)
in order to compare the accuracy of the developed model.
Table 5.7: Results comparison for angular displacement, velocities & accelerations
obtained by the developed program performed in MathCAD and Working Model Software
Angular: displacement, velocities, and accelerations

Body Wheel position Developed program Working Model Software
centers θ(t) ω(t) ε(t) θ(t) ω(t) ε(t)
[deg] [s-1] [s-2] [deg] [s-1] [s-2]
zMmin (t = 0 s) 249.097 0.000 21.390 249.109 0.000 20.535
1
zMrest (t = π/2 s) 260.600 2.744 -16.495 260.600 2.748 -16.494
C1
zMmax (t = π s) 254.704 0.000 15.645 254.711 0.006 15.633
zMmin (t = 0 s) 209.174 0.000 -30.547 209.156 0.000 -29.540
2
zMrest (t = π/2 s) 182.957 -24.070 1.559 182.963 -24.070 1.565
C2
zMmax (t = π s) 159.083 0.000 23.937 159.094 0.010 23.935
- 78 -
zMmin (t = 0 s) 335.184 0.000 17.778 335.195 0.000 17.413

3
zMrest (t = π/2 s) 355.398 21.348 1.388 355.393 21.347 1.383
C3
zMmax (t = π s) 378.210 0.000 -25.849 378.198 -0.011 -25.841
zMmin (t = 0 s) 270.000 0.000 0.000 270.000 0.000 0.000
4
zMrest (t = π/2 s) 270.000 0.000 0.000 270.000 0.000 0.000
C4
zMmax (t = π s) 270.000 0.000 0.000 270.000 0.000 0.000
zMmin (t = 0) 249.097 0.000 21.390 249.109 0.000 20.535
5
zMrest (t = π/2 s) 260.600 2.744 -16.495 260.600 2.748 -16.494
C5
zMmax (t = π s) 254.704 0.000 15.645 254.711 0.006 15.633
zMmin (t = 0 s) 209.174 0.000 -30.547 209.156 0.000 -29.540
6
zMrest (t = π/2 s) 182.957 -24.070 1.559 182.963 -24.070 1.565
C6
zMmax (t = π s) 159.083 0.000 23.937 159.094 0.010 23.936
zMmin (t = 0 s) 270.000 0.000 0.000 270.000 0.000 0.000
7
zMrest (t = π/2 s) 270.000 0.000 0.000 270.000 0.000 0.000
C7
zMmax (t = π s) 270.000 0.000 0.000 270.000 0.000 0.000
Table 5.8: Results comparison for length, relative velocity & acceleration of the spring
& shock absorber, obtained by developed program and Working Model Software
Length, relative velocity & acceleration of spring & damper

Wheel Developed program Working Model Software
Body
position Ls(t) vLs(t) aLs(t) Ls(t) vLs(t) aLs(t)
[mm] [mms-1] [mms-2] [mm] [mms-1] [mms-2]
Spring & zMmin (t = 0 s) 483.527 0.000 -56.902 483.492 0.000 -55.826
shock zMrest (t = π/2) 414.556 -76.455 -10.499 414.576 -76.453 -10.483
absorber zMmax (t = π s) 331.509 0.000 93.207 331.552 0.038 93.182
Table 5.9: Results comparison for total and projection in y and z-axis of velocities for
mechanism components, obtained by developed program and Working Model Software
Velocities of the components of the suspension mechanism

Body Wheel position Developed program Working Model Software
centers vy(t) vz(t) v(t) vy(t) vz(t) v(t)
[mms ] [mms ] [mms ] [mms ] [mms ] [mms-1]
-1 -1 -1 -1 -1
zMmin (t = 0 s) 0.000 0.000 0.000 0.000 0.000 0.000

1
zMrest (t = π/2 s) 5.198 -0.860 5.268 5.206 -0.862 5.276
C1
zMmax (t = π s) 0.000 0.000 0.000 0.012 -0.003 0.012
zMmin (t = 0 s) 0.000 0.000 0.000 0.000 0.000 0.000
2
zMrest (t = π/2 s) 3.893 124.140 124.201 3.896 124.138 124.138
C2
zMmax (t = π s) 0.000 0.000 0.000 0.005 -0.054 0.054
zMmin (t = 0 s) 0.000 0.000 0.000 0.000 0.000 0.000
3
zMrest (t = π/2 s) 3.139 38.995 39.121 3.142 38.994 39.121
C3
zMmax (t = π s) 0.000 0.000 0.000 0.006 -0.018 0.019
zMmin (t = 0 s) 0.000 0.000 0.000 0.000 0.000 0.000
4
zMrest (t = π/2 s) 10.395 -1.721 10.537 10.411 -1.724 10.553
C4
zMmax (t = π) s 0.000 0.000 0.000 0.024 -0.006 0.019
zMmin (t =0 s) 0.000 0.000 0.000 0.000 0.000 0.000
5
zMrest (t = π/2 s) 5.198 -0.860 5.268 5.206 -0.862 5.276
C5
zMmax (t = π s) 0.000 0.000 0.000 0.012 -0.003 0.012
6 zMmin (t = 0 s) 0.000 0.000 0.000 0.000 0.000 0.000
C6 zMrest (t = π/2 s) 3.893 124.140 124.201 3.896 124.138 124.199
- 79 -
zMmax (t = π s) 0.000 0.000 0.000 0.005 -0.054 0.054

zMmin (t = 0 s) 0.000 0.000 0.000 0.000 0.000 0.000
7
zMrest (t = π/2 s) -2.608 250.000 250.014 -2.620 250.000 250.014
C7
zMmax (t = π s) 0.000 0.000 0.000 -0.013 -0.102 0.103
Table 5.10: Results comparison for total and projection in y and z-axis of acceleration
for mechanism components, obtained by developed program and Working Model Software
Acceleration of the components of the suspension mechanism

Body Wheel Developed program Working Model Software
centers position ay(t) az(t) a(t) ay(t) az(t) a(t)
[mms ] [mms ] [mms ] [mms ] [mms ] [mms-2]
-2 -2 -2 -2 -2
zMmin (t = 0 s) 38.363 -14.652 41.066 36.891 -13.902 39.424

1
zMrest (t = π/2 s) -31.202 5.421 31.669 -31.199 5.422 31.667
C1
zMmax (t = π) 28.973 -7.924 30.037 28.951 -7.914 30.013
zMmin (t = 0 s) -1.240 110.348 110.355 -0.757 107.721 107.724
2
zMrest (t = π/2 s) -9.109 5.421 10.600 -9.106 5.396 10.585
C2
zMmax (t = π s) 13.200 -132.924 133.578 13.182 -132.901 133.553
zMmin (t = 0 s) 13.674 29.571 32.579 13.287 29.014 31.912
3
zMrest (t = π/2 s) -14.325 3.704 14.796 -14.323 3.697 14.793
C3
zMmax (t = π s) 14.803 -44.998 47.371 14.789 -44.987 47.356
zMmin (t = 0 s) 76.726 -29.304 82.132 73.782 -27.804 78.847
4
zMrest (t = π/2 s) -62.403 10.842 63.338 -62.399 10.843 63.334
C4
zMmax (t = π s) 57.946 -15.848 60.074 57.901 -15.828 60.026
zMmin (t = 0) 38.363 -14.652 41.066 36.891 -13.902 39.424
5
zMrest (t = π/2) -31.202 5.421 31.669 -31.199 5.422 31.667
C5
zMmax (t = π) 28.973 -7.924 30.037 28.951 -7.914 30.013
zMmin (t = 0 s) -1.240 110.348 110.355 -0.757 107.721 107.724
6
zMrest (t = π/2 s) -9.109 5.421 10.600 -9.106 5.396 10.585
C6
zMmax (t = π s) 13.200 -132.924 133.578 13.182 -132.901 133.553
zMmin (t = 0 s) -79.206 250.000 262.247 -75.296 243.247 254.634
7
zMrest (t = π/2 s) 44.185 0.000 44.185 44.187 -0.051 44.187
C7
zMmax (t = π s) -31.546 -250.000 251.982 -31.537 -249.974 251.955
From Tables 5.7 to 5.10 for the three position of motion at the center of the wheel
rebound, rest and bound, it can be observed that, the obtained results by commercial Working
Model Software coincided very well with results obtained by the developed program
performed in MathCAD environments. The results are practically identical. This verifies the
reliability of the developed kinematic modeling of the suspension mechanism.
Figure 5.19 presents the kinematic results obtained by Working Model Software when
the center of the wheel is in rest position or let us say when the vehicle is in a static position.
After setting all necessary kinematic characteristic for the suspension mechanism, now
it is possible to determine the unknown reaction forces in each joint of the suspension
mechanism by employing expression (A.33). It is supposed that, the maximal tire loads for
one wheel during bump motion will be up to 3g according to authors Blundell and Harty [59].
- 80 -
Rest position
Figure 5.19: Kinematic results obtained in Working Model Software
According to these authors, the maximal tire force that acts on one wheel in vertical direction
is Fw = 8 461 N, whilst forces in other direction are considerably smaller. By employing force
Fw = 8 461 N and the necessary data tabulated in Table 5.6, the spring force Fs(t), the shock
absorber damping force Fsh(t) and reaction forces Fj(t) for joints; A, B, C2, C4, D4, D6, E, F, G,
H and I are presented by the diagrams in Figures A.4 to A.14 as a function of times.
5.4 Optimization of the suspension mechanism
The final design of the suspension mechanism was obtained from the initial design concept
because it is clearly evident that this design ensures suitable performance for the developed
terrain vehicle. Therefore, some of the dimensions of the mechanism can be still changed. The
idea is to determine optimal dimensions of mechanism in order for the suspension mechanism
to reduce as much as possible the lateral displacement of the wheels known as track changes.
This optimization of the dimensions of the terrain vehicle suspension mechanism will
improve the stability of the vehicle when is driven around corners and on uneven road.
5.4.1 The problem of optimum design
Over past years, optimization of mechanical systems has become almost an obligatory part of
the design process. During design processing, the difficulties of the optimization process may
vary over a wider range. On the one hand, there are the simplest problems with linear
response, where it is necessary to optimize only a few simple parameters. In the other hand,
- 81 -
complex problems exist with highly non-linear responses that require lots of effort. The
optimization of the mechanical systems is often based on the non-linear mathematical
programming problem F, determined by the following form:
min f 0 (b, u) , (5.63)

subjected to the constraint:
h j ( b, u) ≤ 0 ; j = 1,L , i , (5.64)
where the scalar functions f0 and hj are termed the objective and constraint function,
respectively. The vector b ∈ Rn denotes design variables, whilst u ∈ Rm is the vector of the
response variables, usually presenting displacement and rotation of the mechanical systems,
discrete by finite elements (Figure 5.20). The symbol n denotes the number of design
variables, m is the number of response variables, and i is the number of constrains. The
objective function is related to the quality of the design, whilst the constraint function usually
reflects the nodal displacement or rotation, design variable limits, technological limitations
and so on, Lenart [52], Kegel et al. [61], Pehan et al. [62], Kegl et al. [63], Kegl and Antes
[64], Kegl and Brank [65] and Zidarič [66].
Therefore, in such an optimization problem the objective and constraint functions are
assumed to be continuous design variables and problem F can be solved efficiently by
utilizing one of the gradient methods which is based on the mathematical programming.
Therefore, the solution procedure for solving the non-linear problem F is an iterative process
and can be described by the following steps:
Step 1: Determining initial values (k = 0) of the design variables b(k) = b(0),

Step 2: Computing f0 and hj, j = 0, ... , i at b(k) (response analysis),
Step 3: Computing df0/db and dhj/db, j = 0, ... , i at b(k) (sensitivity analysis),
Step 4: Submission of the calculated values to the optimization algorithm, in order to
obtain some improvement in design variables ∆b(k) and computing improved
design variables b(k+1) = b(k) + ∆b(k),
Step 5: Check convergence criteria - if it fulfilled exit, otherwise go back to step 2.
Objective function f0 and constraint function hj are in relation to design b and response
variables u. However, before any optimization process is determined the response variable u
at given design variable b needs to be dealt with. In other words, the dependency u = u(b) is
established implicitly by using the equilibrium equation of the structure in general form, as:
F − R = 0, (5.65)
- 82 -
where F and R denote the vectors of the structural internal and external forces, respectively.
The response variable u(k) is obtained from equilibrium equation (5.65), whilst (du/db)(k) has
to be determined from the sensitive equation (more detail is given in literature [65]). These
derivatives are commonly expressed by differentiation of the equation (5.65) with respect to
design variables b:
∂F du ∂F  ∂R du ∂R 
⋅ + −  ⋅ +  = 0 , or (5.66)
∂u db ∂b  ∂u db ∂b 
−1
du  ∂F ∂R   ∂R ∂F 
= −  − . (5.67)
db  ∂u ∂u   ∂b ∂b 
The inverse matrix [∂ F/ ∂ u − ∂R/ ∂u ]−1 is a tangential stiffness matrix of the structure.
This matrix is already known from the response analyses. Equation (5.67) can be solved with
a rather small additional computer effort: only partial design derivatives of internal and
external forces (∂R/∂b − ∂F/∂b) need to be computed.
5.4.2 The modeling of the suspension mechanism by finite elements
The full terrain vehicle suspension system for the front and rear axles are modeling by Finite
Elements performed in STAKx [67] software presented in Figure 5.20.
a) z b) z
[mm] [mm] F
A
B C
G
x D E
[mm] CG -x -y CG y
[mm]
c) d) z
-y
CG
x
y
[mm]A, B A’, B’ CG -x
C, D C’, D’
E E’, F
G x
y
[mm]
Figure 5.20: Modeling of terrain vehicle suspension system by Finite Elements in STAKx
software; a) side view, b) front view, c) plane view, and d) 3D view
- 83 -
Figure 5.21 present the analyzing of the terrain vehicle suspension

suspension system when on the
right wheels of the front and rear axles vertical tire force is applied by Fw = 8461 [N] causing
bound position of wheel ZMmax = 250 [mm]. Whilst, the left wheel of the front and rear axles is
in rebound position ZMmin = -250
250 [mm].
Figure 5.21: Analyzing of vertical wheel motion by Finite Elements in STAKx [67].
After modeling of the terrain vehicle suspension system in STAKx,, the environments
ensure conditions for the optimization of the suspension mechanism caused by applied forces.
5.4.3 The optimal design of the suspension mechanism
In order to obtain an optimal design of the suspension mechanism, the iGOx [68] optimization
program was employed, as is shown in Figure 5.22. This is a separate (stand
(stand–alone)
optimization program containing
aining the approximation method described by Pehan et al. [62],
Kegl et al. [63], Kegl and Antes [64], Kegl and Brank [65] and Zidarič
Zidari [66].
Initial design iGOx optimization program Improve design

variables b(0) (optimizer) variables b(k+1)
b(k)
RUN ACQUIRE
simulators output data
STAKx FE analysis program

(simulators)
Figure 5.22: The optimization procedure for obtaining optimal design variables
- 84 -
Essentially, iGOx run simulators in order to get the values of the objective and
constraint functions from the FEA program (simulators) called STAKx. After that it calls its
own built-in optimization algorithm to improve the values of the design variables. This
procedure is repeated in each cycle of the optimization process, as is presented in Figure 5.23.
The objective and constraint functions are derived in that way, the lateral displacement
of the wheel during bound and rebound wheel motion to be as small as possible. The design
variables are represented by joints position in y and z – axis. These positions of the joints
connection are required to optimize. Table 5.11 presents the position of joints connection
represented by design variables which present the distance of the joint shown in Figure 5.20.
Table 5.11: Position of joints connection and design variables
Joints point x - axis y - axis z - axis

A 1515 253 + 20b1 297+20b2
B 1515 253 + 20b3 127+20b4
A’ 945 253 + 20b5 297+20b6
B’ 945 253 + 20b7 127+20b8
C 1515 331 + 20b9 91.5+20b10
D 1515 331 + 20b11 -78.5+20b12
C’ 945 331 + 20b13 91.5+20b14
D’ 945 331 + 20b15 -78.5+20b16
E 1326 493 + 20b17 -1 + 20b18
E’ 1286 493 + 20b19 -1 + 20b20
G 1326 688 + 20b21 87 + 20b22
F 1286 534 + 20b23 481 + 20b24
Figure 5.23: Optimization procedure performed in iGOx environment
- 85 -
Figure 5.23 shows the optimization process performed in iGOx environment. In Table
5.12 are tabulated the initial and optimal values of the design variables.
Table 5.12 Initial and optimal values of design variables
Design Design
Index Initial Optimal Index Initial Optimal
variable variable
1. b1 1 0.78433204 13. b13 1 0.99150173
2. b2 1 0.88732740 14. b14 1 0.96843082
3. b3 1 0.99612486 15. b15 1 0.83479659
4. b4 1 0.98827393 16. b16 1 0.85650417
5. b5 1 0.84338752 17. b17 1 0.95183656
6. b6 1 0.83591498 18. b18 1 0.88955777
7. b7 1 0.98954070 19. b19 1 0.90502150
8. b8 1 0.99993655 20. b20 1 0.68099104
9. b9 1 0.95295080 21. b21 1 0.74739873
10. b10 1 0.98487951 22. b22 1 0.93353712
11. b11 1 0.88722175 23. b23 1 0.98966755
12. b12 1 0.89947748 24. b24 1 0.89172519
Figure 5.24 shows the wheel path for the initial and optimal design variables when the
positions of the wheel are in bound and rebound resulting from applied forces in the wheel.
300
Initial design
Optimal design
200
ZM
100
0 YM
-100
w/2
-200 LATERAL MOTION
-300
-1055 -1050 -1045 -1040 -1035 -1030 -1025 -1020 -1015
Figure 5.24: Path of wheel for initial and optimal design variables of the terrain vehicle
suspension mechanism
- 86 -
5.5 Discussion of the results

Figures 5.9 and 5.10 show that for minimal input angle θ3min = 335.18 deg, the vertical and
lateral motions of the wheel are ZM = -250 mm and YM = -1017.5 mm - rebound, for rest input
angle θ3rest = 355.40 deg are ZM = 0 mm and YM = -1050 mm - rest, whilst for maximal input
angle θ3max = 378.22 deg are ZM = +250 mm and YM = -1033 mm – bound.
Simulation results performed in MathCAD environment stated that the camber angle during
all vertical motions of the wheel (bound and rebound) was zero: γ (θ3) = 0 deg, kingpin
inclination angle is σ = 12.28 deg, while kingpin offset is: 0.9 mm. These obtained results
help for offering suitable space for fitting brakes, steering gears, and not provide turning
moments in steering wheels during the braking or acceleration processes.
Tables 5.4 presented results for angular displacements of main links, chamber angle,
and the position of the center of the wheel as a function of the input angles of the suspension
mechanism. Tables 5.7 to 5.10 introduced the results for angular displacements, velocities and
accelerations, as well as absolute and projection velocities and accelerations as a function of
ground excitation for the main part of the suspension mechanism. The obtained results are
taken by two ways in order to compare the accuracy of the developed model. One is from the
developed program performed in MathCAD environments, whilst the other by simple
commercial Working Model software. The results are practically identical and coincide very
well with the two methods. This verifies the reliability of the developed kinematic modeling
for the terrain vehicle suspension mechanism.
Appendix A shows the results by diagrams regarding unknown reaction forces for joints
of the suspension mechanism for maximal tire force Fw = 8461 N that could be acts in one
wheel. These reaction forces are obtained from the dynamic modeling of suspension
mechanisms as a function of ground excitation which changes over time.
From Figure 5.24 it can be concluded that, the optimized suspension system offers 0%
rebound, 18% rest and 37.5% bound less lateral wheel displacement when compared with the
initial rebound position.
From discussing the results it can be concluded that, the developed suspension
mechanism allows motion of wheel upwards and downward by providing optimal lateral
wheel displacements and zero camber angle.
- 87 -
6 DESIGNS AND OPTIMIZATIONS OF PASSIVE, ACTIVE

AND SEMI-ACTIVE SUSPENSION SYSTEMS FOR A
TERRAIN VEHICLE
This chapter introduces designs and optimization procedures for passive, active, and semi-
active linear and non-linear suspension systems regarding a terrain vehicle. Due to the
specific requirements for a terrain vehicle such as large vertical motion of the wheels,
comfort, and safety, the objective of this approach is the ability to design, optimize and
analyze the vehicle’s suspension performance. In this way it is possible to decide which kind
of the suspension system resulting from passive, active or semi-active fulfills, as far as
possible, the specific requirements for a terrain vehicle when designing it.
This vehicle’s suspension system is first represented using mathematical modeling
regarding a quarter of it, and the equations of motion are derived at and solved by using
MATLAB/Simulink. In order to verify the reliability of the derived at computer program, a
comparison is made between the measured experimental data. The decision parameters for
passive linear suspension systems are optimized by applying the Multi-Objective Genetic
Algorithms. On the other hand, the parameters for the active damping device are optimized by
using the Hooke-Jeeves method, which is based on non-linear programming. The vehicle’s
suspension performance prior to development is evaluated using initial and optimal designs.
The applications of active and semi-active systems on a concrete terrain vehicle are
presented and compared with the passive systems by evaluating results regarding the vehicle’s
body acceleration, velocity, displacement, and vertical tire force, namely those aspects that
have a directly influence on driving comfort, and driving safety. The results are shown by
diagrams that represent the sensitivities of responses depending on the ground excitation and
other parameters. The driving comfort is related to the vehicle’s vertical body acceleration.
The driving safety is dependent on the quality of the contact between the tires and the ground
surface, and so the wheels should remain in contact with the ground’s surface as firmly as
possible, Belingardi and Demic [69]. Designers always devote particular attention to the
suspension systems in order to improve both characteristics; comfort and safety.
The vehicle suspension systems are categorized as passive, active, and semi-active
systems. In this thesis attention is focused on the active and semi-active systems. These
systems ensure better performances (driving comfort and driving safety) than passive
- 88 -
suspension systems. The key problem regarding the active and semi-active suspension
systems is when determining the optimal active damping force. This force depends on the
operational conditions that come from the sensors. The control programming unit (CPU)
controls the actuator or variable shock-absorber for producing an adequate damping force.
The actuator or variable shock- absorber needs to respond quickly and precisely. The focus of
this approach is a mathematical model of the active damping force and its optimization. The
optimization process for design variables is carried out by using the stochastic parametric
optimization method, Pajaziti [23] and Demic et al. [70]. The optimal design parameters are
obtained when the objective function obtains minimal values. One of the well-established
models for understanding and explaining vehicle suspension is the so-called quarter vehicle
model (QVM), Yu et al. [37] and Ram Mohan Rao et al. [71]. However, for more detailed
analyses more complex models such as half and full vehicle models can also be used. QVM is
comparatively easy to transform into a mathematical model that actually consists of two
second-order differential equations that can be solved in numerical ways.
Due to simplicity it is assumed that the spring and tire stiffness coefficients remain at
constant values. At this stage a powerful computing program is needed to calculate the
suspension’s behavior and its characteristics. It is supposed that considering only the vertical
ground excitation provides usable results. Any eventual longitudinal and lateral excitations
that may arise during vehicle movement are therefore neglected.
6.1 Mathematical modeling of the quarter vehicle suspension system
The quarter-vehicle model consists of two masses, two springs, one or two shock- absorbers,
and probably an actuator, Figures 6.1 and 6.2. The lower mass is the un-sprung mass mu, and
the upper one is the sprung mass ms. The un-sprung mass represents one wheel assembly
mass, whilst the sprung mass represents approximately ¼ of the remaining total vehicle mass.
The lower spring is described by the linear kt or non-linear kt1, kt2, kt3 tire stiffness
coefficients, whilst the shock absorber is by the ct tire damping coefficient. The tire damping
is often neglected due to its insignificant influence on the final results, Jazar [9]. The
suspension system is completed when the spring and the shock absorber are inserted between
two masses. The spring is defined by the linear ks or non-linear ks1, ks2 spring stiffness
coefficients and the shock-absorber is determined by the linear csh or non-linear csh1, csh2
shock absorber damping coefficients. The lower or un-sprung mass is excited by the ground
surface zr through the tires’ contact. Forces that act within the suspension systems can be
- 89 -
described by linear or non-linear characteristics (Figures 6.1 and 6.2). Based on experience,
the real behavior is better described using the non-linear characteristics. In the continuation
the mathematic models are detailed for passive, active, and semi-active suspension systems.
6.1.1 Passive linear and non-linear suspension systems
The linear and non-linear models of the passive quarter-vehicle used for simulation of the
terrain vehicle’s suspension, are shown in Figure 6.1.
Fis
Sprung zs, δzs
Fs Fsh mass (ms)
Linear ks csh Linear
or ks1 csh1 or
Non-linear k csh2 Non-linear
Fiu s2
Fs Fsh Un-sprung zu, δzu
mass (mu)
Ft
kt1
Non-linear kt2 kt Linear
kt3 zr
Figure 6.1: The passive linear and non-linear quarter-vehicle models
For the presented QVM the differential equations of the motion are determined by
applying Lagrange’s equations:
d  ∂E k  ∂E k δ W d  ∂E k  ∂ E k δW
  − = ;   − = . (6.1)
dt  ∂z&s  ∂z s δz s dt  ∂z&u  ∂z u δz u
The kinetic energy Ek is defined by the following expression:
1 1
Ek = ⋅ ms ⋅ z&s2 + ⋅ mu ⋅ z&u2 . (6.2)
2 2
The derivatives of Lagrange’s equations can be written as follow:
d  ∂E k  ∂E d  ∂E k  ∂E
  = ms ⋅ &z&s , k = 0 ;   = mu ⋅ &z&u , k = 0 . (6.3)
dt  ∂z& s  ∂z s dt  ∂z&u  ∂z u
The elementary work δW is given by the following expression:
δW = −(Fs− l or n + Fsh−l or n )⋅ δzs + (Fs− l or n + Fsh−l or n − Ft−l or n )⋅ δzu . (6.4)
The elementary work during virtual displacement of the sprung mass δzs, and un-sprung
mass δzu, is expressed as:
δW δW
= −(Fs−l or n + Fsh−l or n ), = Fs − l or n + Fsh − l or n − Ft − l or n . (6.5)
δzs δzu
- 90 -
The linear forces that act within the suspension system, such as; dynamic tire force Ft-l,
spring force Fs-l, and shock-absorber damping force Fsh-l, are determined by the following
expressions:
Ft − l = k t ⋅ (zu − z r ), (6.6)
Fs − l = k s ⋅ (z s − zu ), (6.7)
Fsh − l = c sh ⋅ (z& s − z&u ). (6.8)
With a few mathematical rearrangements, the Lagrange’s equations of motion for
passive linear suspension systems can be written as follows:
ms ⋅ &z&s = −ks ⋅ (zs − zu ) −csh ⋅ (z&s − z&u ),

(6.9)
mu ⋅ &z&u = ks ⋅ (zs − zu ) + csh ⋅ (z&s − z&u ) − kt ⋅ (zu − zr ).
In matrix form the equation (6.9) is written as follows:
ms 0  &z&s  − csh csh  z&s  − ks ks  zs   0 

 0 m  ⋅ &z&  =  c ⋅  +  ⋅   +   ⋅ zr . (6.10)
 u   u   sh − csh  z&u   ks − (ks + kt ) zu  kt 

The non-linear behaviors of those dynamic forces that act within the suspension are
high-order polynomial functions. The non-linear dynamic tire force Ft-n is described as a
third-order polynomial function, spring force Fs-n is given by a third-order polynomial
function, whilst the shock-absorber damping force Fsh-n is modeled as a second-order
polynomial function.
These forces are determined by the following expressions, Pajaziti [23] and Demic [72]:
2 3
Ft −n = kt1 ⋅ (zu − zr ) + kt 2 ⋅ (zu − zr ) − kt 3 ⋅ (zu − zr ) , (6.11)
3
Fs−n = ks1 ⋅ (zs − zu ) + ks2 ⋅ (zs − zu ) , (6.12)
2
Fsh−n = csh1 ⋅ (z&s − z&u ) + csh2 ⋅ (z&s − z&u ) sign(z&s − z&u ). (6.13)
The differential equations of motion for the passive non-linear suspension systems are:
3 2
ms ⋅ &z&s = −ks 1 ⋅ (zs − zu ) − ks 2 ⋅ (zs − zu ) − csh 1 ⋅ (z&s − z&u ) − csh 2 ⋅ (z&s − z&u ) sign (z&s − z&u ),
3 2
mu ⋅ &z&u = ks 1 ⋅ (zs − zu ) + ks 2 ⋅ (zs − zu ) + csh 1 ⋅ (z&s − z&u ) + csh 2 ⋅ (z&s − z&u ) sign (z&s − z&u )K (6.14)
2 3
− kt 1 ⋅ (zu − zr ) − kt 2 ⋅ (zu − zr ) + kt 3 ⋅ (zu − zr ) ,
where &z&s , z& s , zs and &z&u , z& u , z u denote the acceleration, velocity, and displacement of the
sprung and un-sprung masses, respectively. Whilst (z s − z u ) , (z u − z r ) and (z& s − z& u ) represent
the suspension travel, tire deflection, and relative velocity of the shock absorber.
- 91 -
Equations (6.9), (6.10) and (6.14) represent a system of second-order non-homogeneous

linear differential equations regarding the motions for linear and non-linear passive
suspension systems, respectively.
6.1.2 Active and semi-active non-linear suspension systems
Many attempts regarding control strategies on active and semi-active suspension systems have
been performed over past years to improve driving comfort and driving safety. Several control
strategies have been developed for the semi-active, such as on-off skyhook, on-off ground-
hook, continuous skyhook, fuzzy logic control, Abramov et al. [73]. Consequently, a CPU-
controlled active system for adjusting damping force has been implemented here.
The main problem concerning the already-mentioned control systems is to determine an
adequate active damping force. Determination of this force is done by the addition of two
virtual passive shock-absorbers. The first one is fixed on the sprung mass whilst the other one
is fixed on the un-sprung mass. The second fixing-point for both shock-absorbers is fixed in
the fictive hook on the sky. The damping force caused by the first virtual shock-absorber,
always acts in the opposite direction to the velocity of the sprung mass, whereas the other
damping force produced by the second virtual shock absorber, always acts in the same
direction as the velocity of the un-sprung mass. In reality, any addition of the skyhook
approach is impossible because hooking the shock-absorber onto the sky is also impossible.
The real implementation of the skyhook approach is possible by using an active actuator
installed between the sprung and un-sprung masses, Figure 6.2, Lajqi and Pehan [2].
Fictive hook in sky

Fis Sensor
bs
Virtual Sprung Sprung zs
shock absorber mass (ms) mass (ms)
Fs Fa cvariable
Fictive hook ks1 csh1 ks1
Fa CPU
in sky ks2 csh2 ks2 Fsh Fa or Fsh
Fiu Sensor
bu
Virtual Un-sprung Fs Fa Un-sprung zu
shock absorber mass (mu) Ft mass ( m u )
kt1 kt1
kt2 kt2
kt3 kt3 zr
IDEAL SKYHOOK SYSTEM REAL SKYHOOK SYSTEM

Figure 6.2: Skyhook concept for active and semi-active non-linear suspension systems
In this case the active damping force is described by the expression:
- 92 -
Fa = −bs ⋅ z&s + bu ⋅ z&u , (6.15)

where bs and bu denote the damping coefficients of both virtual shock-absorbers. The
adequate damping coefficients bs and bu are determined by an optimization process.
The active damping force Fa given by equation (6.15) can be re-written as follows:
 b 
Fa = bu ⋅ ( z& u − z& s ) − b s  1 − u  ⋅ z& s , (6.16)
 bs 
where the first member, (z&u − z&s ) of the equation (6.16) relates to the relative velocity of the
active actuator, whilst the second member relates to the absolute velocity of the sprung mass.
In order to fulfill, as far as possible, the specific requirements given in this thesis
dealing with active suspension systems, the priority is dedicated to hydraulic active actuators.
For a semi-active suspension system dealing with variable shock-absorber, regulation of the
damping characteristic by using servo-valve will be considered.
The dynamic equations of motion for the active non-linear suspension system derived
from Figure 6.2, are written as follow:
3
m s ⋅ &z&s = − k s1 ⋅ (z s − z u ) − k s 2 ⋅ (z s − zu ) + bs ⋅ z& s − bu ⋅ z&u ,
3
mu ⋅ &z&u = k s1 ⋅ (z s − zu ) + k s 2 ⋅ (z s − zu ) − bs ⋅ z& s + bu ⋅ z&u L (6.17)
2 3
− k t 1 ⋅ ( z u − z r ) − k t 2 ⋅ (z u − z r ) + k t 3 ⋅ (z u − z r ) .
The dynamic equations of motion for the semi-active non-linear suspension systems
determined from Figure 6.2, generate the following form:
ms ⋅ &z&s = − k s1 ⋅ (z s − zu ) − k s 2 ⋅ (z s − zu ) + bs ⋅ z&s − bu ⋅ z&u , 

3
3

mu ⋅ &z&u = k s1 ⋅ (z s − zu ) + k s 2 ⋅ (z s − zu ) − bs ⋅ z&s + bu ⋅ z&u L IF (Fa ⋅ Fsh ) ≤ 0 − ON
2 3 
− k t1 ⋅ (zu − z r ) − k t 2 ⋅ (zu − z r ) + k t 3 ⋅ (zu − zr ) . 
(6.18)
ms ⋅ &z&s = − k s1 ⋅ (z s − zu ) − k s 2 ⋅ (z s − zu ) − csh1 ⋅ (z&s − z&u )L 
3
2

− csh 2 ⋅ (z&s − z&u ) sign (z&s − z&u ), 
3 
mu ⋅ &z&u = k s1 ⋅ (z s − zu ) + k s 2 ⋅ (z s − zu ) + csh1 ⋅ (z&s − z&u )L  IF (Fa ⋅ Fsh ) > 0 − OFF
2 
+ csh 2 ⋅ (z&s − z&u ) sign (z&s − z&u ) 
− k t1 ⋅ (zu − z r ) − k t 2 ⋅ (zu − z r ) + k t 3 ⋅ (zu − zr ) . 
2 3
Forces that act within active and semi-active suspension systems are taken for non-
linear behavior, as determined by equations (6.11), (6.12), and (6.13).
The ON-OFF skyhook control given in equation (6.18) presents an effective vibration
control strategy. The CPU-respond IF (Fa ⋅ Fsh ≤ 0) causes shock-absorber adjustment in the
high-damping state. The CPU-respond IF (Fa ⋅ Fsh > 0) causes the shock absorber adjustment
- 93 -
in the low-damping state. Adjusting the shock-absorber within high or low states depends on
the product regarding the relative speed of the shock-absorber (z& s − z&u ) and the absolute
speeds of the sprung and un-sprung masses ( z& s and z& u ). If this product is zero or negative, the
shock-absorber is adjusted to a higher-damping state, otherwise it is set to a low-damping
state.
Equations (6.17) and (6.18) represent a system of second-order non-homogeneous,
non-linear differential equations regarding the motions for non-linear active and semi-active
suspension systems, respectively.
6.2 Simulations of vehicle suspension by numerical methods
In order to solve the differential equations of motion for the general case of ground excitation,
it is more than necessary to have an efficient tool to speed-up the numerical procedures.
Numerical simulations have been developed for all the presented suspension systems. A
methodology is presented for solving differential equations of motion for passive, linear, and
semi-active non-linear suspension systems. Whereas for other systems the procedure is the
same, here only the output results are presented, as obtained from simulation. In order to
speed-up the numerical procedures, the differential equations of motion need to be
transformed into more suitable forms.
Equation (6.9) is transformed into the state variables equation by simplifying the second
order differential equations into first-order differential equations. It is supposed that the state
variables are given by the following expressions:
x1 = zs , x2 = zu , x3 = z&s , x4 = z&u ,
(6.19)
dx1 = x3 = z&s , dx2 = x4 = z&u , dx3 = &z&s , dx4 = &z&u .
Substituting equation (6.19) into equation (6.9), the state variables equation is written as
follows:
dx1   0 0 1 0   x1   0 
dx   0 0 0 1   x2   0 
 2  ⋅ +   ⋅ {zr }.
 = (6.20)
dx
 3  − k s / ms k s / ms − csh / ms csh / ms  x
 3  0 
dx4   ks / mu − (ks + kt ) / mu csh / mu − csh / mu   x4  kt / mu 
  
The state variables for the semi-active non-linear suspension system derived from
equation (6.19), are expressed as follow:
- 94 -
dx1 = x3 = z&s ,
dx2 = x4 = z&u ,
If (Fa ⋅ Fsh ) ≤ 0 − ON ( ACTIVE − HIGH STATE )
3
dx3 = &z&s = (1/ ms ) ⋅ {−ks1 ⋅ (x1 − x2 ) − ks2 ⋅ ( x1 − x2 ) + bs ⋅ x3 − bu x4 },
3
dx4 = &z&u = (1 / mu ) ⋅ {ks1 ⋅ ( x1 − x2 ) + ks 2 ⋅ ( x1 − x2 ) − bs ⋅ x3 + bu ⋅ x4 L
2
− kt1 ⋅ (x2 − zr ) − kt 2 ⋅ (x2 − zr ) + kt 3 ⋅ (x2 − zr )}. (6.21)
If (Fa ⋅ Fsh ) > 0 − OFF (PASSIVE − LOW STATE )

3 2
dx3 = &z&s = (1/ ms ) ⋅ {−ks1 ⋅ (x1 − x2 ) − ks2 ⋅ ( x1 − x2 ) − csh1 ⋅ ( x3 − x4 ) − csh2 ⋅ ( x3 − x4 ) sign (x3 − x4 )},
3
dx4 = &z&u = (1 / mu ) ⋅ {ks1 ⋅ ( x1 − x2 ) + ks 2 ⋅ ( x1 − x2 ) + csh1 ⋅ ( x3 − x4 )K
2 2 3
+ csh2 ⋅ ( x3 − x4 ) sign (x3 − x4 ) − kt1 ⋅ ( x2 − zr ) − kt 2 ⋅ ( x2 − zr ) + kt 3 ⋅ ( x2 − zr ) }.
Equation (6.20) represents the general form of the state variables equation for passive
linear suspension system in matrix form, whilst equation (6.21) defines the state variables
equation for a semi-active, non-linear suspension system.
The flowchart diagram for solving differential equations of motion is shown in
Figure 6.3. The program starts by reading the suspension parameters, ground excitation,
number of input and output variables, the initial START
conditions, and the number of iterations (time). Then the Number of
program continues to solve differential equations of Suspension variables
parameters Number of
motion by applying the solver ode45 in a iteration
Initial
MATLAB/Simulink environment. The program stops conditions Ground
excitation
when the iteration condition is fulfilled. The results are
Runge-Kutta
presented through diagrams. The output results,
MATLAB/Simulink for
depending on time, are the accelerations, velocities, and solving differential equations
of motion by numerical ways
displacements of the sprung mass. The program also
computes the vertical tire forces, which act between the NO t ≥ ti+∆t
ground’s surface and the tire. The output results are
within the function of ground excitation that represents YES
RESULTS:
road profile or so-called road bumps. In this thesis the - Ground excitation,
- Accelerations,
ground excitation is demonstrated by two repeated - Velocities,
smooth obstacles on a flat road, which is approximated - Displacements, and
- Vertical tire forces.
by a smooth function, such as the cosine function shown
STOP
in Figure 6.4.
Figure 6.3: Flowchart diagram for solving differential equations of motion
- 95 -
zr(t)
H1
H2
v1, t1 v1, t2 v2, t3 v2, t4 v2, t5
x1 x2 x3 x4 x5 xi(t)
Figure 6.4: Ground excitation represented by double cosine road bumps
The ground excitation zr(t) is written, as follow:

0.5 ⋅ H1 ⋅ [1 − cos (8π ⋅ t )] if t1 ≤ t ≤ t2

zr (t ) = 0.5 ⋅ H2 ⋅ [1 − cos (4π ⋅ t )] if t3 ≤ t ≤ t4 (6.22)
0 else ,

where H1, H2 are amplitudes of the first and second bumps, v1, v2 denotes vehicle speed,
xi define each part of road, and ti is the termed time for passing each part of road. Table 6.1
gives the appropriate data for simulation of the ground excitation computed by expression
(6.22).
Table 6.1: Description data for road bumps when the vehicle passed over them
Road data Amplitudes of the road bumps: H1 = 0.05 [m], H2 = 0.1 [m]
i xi [m] for v1 = 20 [kmh-1], ti =xi/v1 [s] for v2 = 10 [kmh-1], ti =xi/v2 [s]
1 2.78 0.50 -
2 1.39 0.25 -
3 3.47 - 1.25
4 1.39 - 0.50
5 4.17 - 1.50
A few authors such as Rill [4] and Sam and Osman [74] have used similar functions for
ground excitation, as described here. Therefore, road bumps introduced by cosine functions,
can be considered as one of the complex ground excitation inputs.
6.3 Verifying the reliability of the developed program
In order to verify the presented computer program, the results are compared to those done in
an experimental ways. The input data are: ms = 320 kg, ks = 15450.42 Nm-1,
csh = 1544.94 Nsm-1, whilst mu = 25 kg and kt = 400000 Nm-1, are adopted. These data are
obtained from experimental methods for a passive suspension system within an Opel Vectra
1.8i passenger car. The experiment is performed at Graz of University Technology, Vehicle
Safety Institute (VSI), Figure 6.5. More details regarding the performed experiment are given
in reference Lajqi et al. [18].
- 96 -
Vehicle's Displacement
body sensor
Force sensor VSI Laborator

Figure 6.5: Laboratory devices for determining suspension characteristics, Lajqi et al. [18]
In order to verify the reliability of the developed program, ground excitation and other
decision parameters are used that corresponded to the one explained above. The ground
excitation is theoretically simulated by step bump as a function of time, Figure 6.6.a. Only the
passive linear suspension system is considered for comparison.
Figure 6.6 show the results in the form of ground excitation and displacements that
correspond to the vehicle’s body (sprung mass), depending on time. These results are obtained
by experimental and simulation methods, respectively. Finally, the displacements of the
vehicle’s body are compared, as obtained by different approaches.
Experiment
0.30 0.30 Simulation
Vehicle body displacement [m]
Ground excitation [m]
0.29 0.29
0.28 0.28
0.27 0.27
0.26 0.26
0.25 0.25
0 2 4 6 8 0 2 4 6 8
a) Time [s] b) Time [s]
Figure 6.6: a) Ground excitation, and b) vertical vehicle body displacements, Lajqi et al. [18]
From Figure 6.6.b it is observed that, the simulation results performed in MATLAB
environment generates sufficient and reliable results when compared to those produced by the
experimental method. Consequentially, it can be concluded that the developed program is
suitable for use during further analysis of the vehicle’s suspension behavior.
- 97 -
6.4 Optimization of the suspension parameters for passive, active, and

semi-active suspension systems
6.4.1 Optimization of the suspension parameters for a passive linear suspension system
In order to ensure the optimal suspension parameters for a terrain vehicle equipped with a
passive linear suspension system, driving comfort and driving safety as design criteria are
taken into consideration. The driving comfort is evaluated by computing the standard
deviation of the vehicle’s body acceleration - , where greater values are undesired. The
driving safety is estimated by the standard deviation of the dynamic tire forces acting between
the ground’s surface and the tires - . If the tire forces oscillate too much then the contact
between the ground surface and the tire is weak. Consequentially the vehicle’s
maneuverability and steering would be in question. In order to obtain optimal suspension
design parameters regarding driving comfort and driving safety, according to Gobbi and
Mastinu [34], Mastinu et al. [46] Lajqi et al. [75], and Popp and Schiehlen [76], the following
expressions are used:
a. The standard deviation of the vehicle’s body acceleration ( - driving comfort):
1
 A ⋅ v  ms mu 2 csh  2
σ &z&s =  b ⋅  2
⋅k 2
s + 2
⋅k s + 2
⋅ k t   . (6.23)
 2 c ⋅m
 sh s c sh ⋅m s m s 
b. The standard deviation of the dynamic tire forces ( - driving safety):
 Ab ⋅ v  (ms + mu )3 2 csh ⋅ (ms + mu )2

σF =  ⋅ ⋅k s+ ⋅ kt + K
 2  csh ⋅m s
2
t
m2s
1 (6.24)
m 2 ⋅ mu ⋅ (ms + mu ) 2
+ u ⋅k 2− t ⋅k s ⋅ k 
t  .
csh ms ⋅ csh 
where Ab denotes the road roughness parameter, whilst v is the vehicle speed.
The equations (6.23) and (6.24) define the design criteria (driving comfort and driving
safety) used as objective functions for optimizing the passive linear suspension parameters.
An optimization technique is used for determining the optimal suspension parameters
for the passive linear suspension system --- the Multi-Objective Genetic Algorithms
(MOGAs) method. In order to explain the optimization procedure by employing MOGAs, the
Genetic Algorithms process is first introduced.
- 98 -
6.4.1.1 Genetic Algorithms
The genetic algorithms (GAs) are well-known as methods for the global optimization of
highly complex problems, which were invented by Professor John Holland at the University
of Michigan in the 1970s, Alkhatib et al. [36]. GAs is a stochastic global search and
optimization method that mimics the metaphor of natural biological evolution and is
developed based on Darwin’s theory of evolution “survivor of the fittest”. Figure 6.7 shows a
flowchart of the evolutionary process as performed by GAs.
Select parents
Parents
Initialization
Crossover
Population
Mutation
Repeat until Offspring
satisfied Fitness
Figure 6.7: GAs evaluation process
The GAs does not use gradient information but rather perform stochastic searches of the
design in order to minimize the values of the objective functions. The objective functions are
not required to be continuous, neither do they need information about their derivates;
therefore they can handle problems with discrete solution spaces. Evaluation of the objective
functions and constraints is performed by fitness. In several aspects, GAs differs from
traditional optimization techniques, Yu and Yu [37] and Zhang [77]. Some of these
differences are summarized as follows:
- GAs are able to solve complex problems using non-convex solution space, where
traditional methods usually fail,
- GAs searching mechanisms are stochastic in nature, which makes them capable of
searching entire solution space with more probability of finding the global solution,
- GAs solutions search from a population, not from a single solution,
- GAs allow design variables to be mixed (continuous and discrete variables),
- GAs use probabilistic transition rules not deterministic rules. This characteristic
makes GAs uniquely suited to MOGAs optimization.
All these described differences make the GAs superior over classic optimization
methods in some real world applications, particularly for complex and sophisticated
engineering problems. The basic procedure of the GAs consists of the following steps Senthil
kumar [19], Yu and Yu [37], and Zhang [77]:
- 99 -
Step 1#: Initial population

GAs start iterations with an initial population (design solution). The initial population of
chromosomes is created rather and consists of a set of design solutions. Each chromosome is
used to express a sequence of genes from the binary digits (0 and 1). Hence, real expressions
are used to represent chromosome for objective functions when considering optimization. The
chromosomes exist within the architecture regarding the design variable’s representation by
the spring and tire stiffness coefficients, shock absorber damping coefficient, and the sprung
and un-sprung masses.
Step 2#: Fitness function
The fitness function evaluates all the chromosomes. The target of the fitness function is
to numerically encode the performance of the chromosomes. The fitness function for the
treated problem consists of an objective function that enables minimization of the vehicle’s
body acceleration and dynamic tire forces.
Step 3#: Selection function
The selection function is one of the key procedures in order to produce successive
generations. It determines which chromosomes (individuals) will be represented and continue
on to the next generation. Selection starts from the old population to the new population based
on the fitness’s of the chromosomes, as given by the evaluation function. Chromosomes with
the greater fitness rates are selected whilst chromosomes with lower fitness rates are removed
from the population. There are several methods for selecting chromosomes, which have been
reported in the literature [78], such as: tournament selection, roulette-wheel selection, ranking
selection, etc.
The tournament selection method, selects each parent by choosing individuals at
random, by a specified number of turnamnet size, and then choosing the best individual out of
that set to be a parent. Thus the tournament selection method will be used as the selection
function during the optimization process.
Step 4#: Genetic operators (crossover and mutation)
Generally speaking, there are two types of genetic operators: crossover and mutation.
The crossover process allows for an exchange of genes (design solution) between individuals
in the population. Crossover takes two individuals (chromosomes) and produces two new
individuals whereas mutation directs an individual to produce a single new solution. If the
parents are allowed to mate, a combination operator is used to exchange genes between the
- 100 -
two parents in order to produce children. If they aren’t allowed to mate, the parents are placed
unchanged into the next generation. Figure 6.8 shows the genetic operators.
The best crossover rate is between 0.80 and 0.95, whilst the best mutation rate for most
applications is between 0.001 and 0.1.
parent 1: 1 1 1 1 1 1 parent: 1 1 1 1 1 1 1
parent 2: 0 0 0 0 0 0
Crossover Mutation
child 1: 1 1 1 0 0 0
child 2: 0 0 0 1 1 1 child: 1 1 1 0 1 1 1
Figure 6.8: Genetic operators (crossover and mutation)
Step 5#: Evaluation function

The evaluation function is used to evaluate the fitness rates of the chromosomes, i.e. the
quality of the solution, within a population. A better solution obtains a higher score. That
function directs the population towards progress because a good solution with a high score
will be selected during the selection process, and a poor solution removed.
Step 6#: Repeated steps 3 - 5
If the termination criterion has not been met, the process of selection, crossover and
mutation (steps 3-5) should be repeated again in order to obtain an optimal design solution.
6.4.1.2 Multi-objective optimization process by using the Genetic Algorithms
In several cases regarding real problems, the multi-objective optimization problem involves
multiple and conflicting objective function. So, optimizing of the design variables with
respect to a single objective function often results in unacceptable results with respect to the
other objective functions. Therefore, a perfect multi-objective solution that simultaneously
optimizes each objective function is almost impossible in many cases, Konak et al. [33]. A
reasonable solution to a multi-objective problem is to investigate a set of solutions. Each set
of solutions should satisfy the objective functions at an acceptable level without being
dominated by any other solution.
GA is one of the suitable techniques for solving multi-objective optimization problems.
The first practical MOGAs, called vector evaluated GAs, was proposed by Schaffer [79].
Afterwards several multi-objective evolutionary algorithms were developed including
MOGAs, Mastinu et al. [46], Fonseca and Fleming [80].
The multi-objective optimization problems are usually written in the following form:
- 101 -
 f1 ( x) 
 f ( x) 
 2 
min f ( x) =  ,
M 
  (6.25)
 f nobj ( x)
 
x = {x1 x2 K xn } ≤ 0 .
T
Subjected to the constraints:
{
h(x) = h1(x) h2 (x) L hnc (x) ≤ 0, }T
(6.26)
where, f(x) is a termed multi-objective function, h(x) denotes the constraint function, vector x
describes the design variables, nobj ≥ 2 gives the number of objective functions, n presents the
number of design variables, whereas nc is the number of constrains.
6.4.1.3 Optimization of the passive linear suspension design parameters
In order to obtain an optimal suspension design for a terrain vehicle with passive linear
suspension system, two design criteria are taken: driving comfort and driving safety. The
driving comfort is presented by the standard deviation of the vehicle’s body
acceleration − σ . The driving safety is determined by the standard deviation of the dynamic
tire forces − . There are five design variables employed during the optimization process
such as spring stiffness coefficient ks, shock-absorber damping coefficient csh, tire stiffness
coefficient kt, sprung ms, and un-sprung mass mu. The objective functions and design variables
limits are written as follows:
 1


 b ⋅
A ⋅ v  m m c   2

s 2
⋅k + u 2
⋅k + sh
⋅k 
  2  c ⋅m 2 s c ⋅m 2 s m 2 t  
  sh s sh s s  
 Ab ⋅ v  (ms + mu )3 2 csh ⋅ (ms + mu )2 
min f ( x) =  ⋅  2
⋅k s + 2
⋅ k t + K (6.27)
 2  c sh ⋅m s ms 
 1 
 mu 2 2 ⋅ mu ⋅ (ms + mu )  2 
 + ⋅k t − ⋅ k s ⋅ k t   
 csh ms ⋅ c sh  
Subjected to the design variables limit:
50000 N / m ≤ ks ≤ 200000 N / m,
5000 Ns/ m ≤ csh ≤ 20000 Ns/ m,
200000 N / m ≤ kt ≤ 500000 N / m, (6.28)
280 kg ≤ ms ≤ 300 kg,
35 kg ≤ mu ≤ 70 kg.
- 102 -
The lower and upper limits of the design variables are selected based on previous
studies. The initial populations of the design variables are selected to be of medium value
between the lower and upper limits. Optimization of the design variables is carried out by
employing MOGAs within the MATLAB environment. In Table 6.2 the used parameters in
MOGAs optimizing process are shown. The optimization process starts by reading the design
variables limit and initial population of the design variables. Then, the design variables pass
for simulation of the differential equations of motion for the passive linear suspension system.
Table 6.2: The MOGAs parameters performed in the MATLAB GAs toolbox
GAs step GAs parameters Values

Design Design variables (constraints)
variables limits Lower limits: [50000 5000 200000 280 35]
Upper limits: [200000 20000 500000 300 70]
Population
Population type: Double vector
Step 1# Population size: 100
Creation function: Uniform
Initial population: [125000 12500 350000 290 50]
Problem
Step 2# Fitness function: ,
Number of variables: 5
Step 3# Selection
Selection function: Tournament
Reproduction
Crossover fraction: 0.8
Mutation
Step 4# Mutation technique: Uniform
Mutation rate: 0.01
Crossover
Crossover function: Heuristic
Crossover rate: 0.8
Stopping criteria
Step 5# Number of generation: 1000
Fitness limit: -∝
Evaluation function (tolerance) 5⋅10-5
After these steps, the design variables are followed within the MOGAs process to
calculate the fitness function. The optimization process will be stopped when the design
variables have been improved and the design criteria is fulfilled, otherwise the optimization
process will be repeated again and stopped whenever the optimal design solution is obtained.
The already described optimization process performed in MOGAs is summarized in the
flowchart diagram shown in Figure B.5. The initial and optimal values of the design variables
- 103 -
for four independent runs by utilizing MOGAs for two objective functions are shown in Table
6.3.
This table gives the values of the objective functions and the number of iterations for
vehicle speed v = 20 kmh-1. During each run the number of iterations is plotted automatically
by computer. It can be observed that, for each new run, significantly similar results for the
design variables are found. The small differences are a result of two conflicting objective
functions, which try to fulfill driving comfort and diving safety at the same time. These
smaller differences can also be noticed in Figure 6.9.b, from the plotting of family points.
Table 6.3: Initial and optimal suspension design parameters
Optimal
Suspension parameters Initial
Run 1 Run 2 Run 3 Run 4
-1
ks [Ns ] 125000 57008 54872 50700 52686
csh [kg] 12500 5686 5360 5170 5073
kt [Ns-1] 350000 202180 207330 208055 204520
ms [kg] 290 298 299 300 295
mu [kg] 50 36 37 35 35
[ms-2] 23.56 12.11 11.87 11.59 11.59
[N] 8057 4158 4139 4049.07 3998
Iteration number - 102 102 102 107
The best design is obtained after the fourth runs, when the evaluation function has met
the design criteria. In this case, the objectives function has reached the minimal values. Figure
6.9, illustrates the convergence of the four runs, by plotting objectives function with single
points, and the average distance between individuals that corresponds to the design variables.
Objective function [N]
a) b)
Average distance
Objective function [ms-2] Generation

Figure 6.9: Optimization process; a) Single point between two objective functions, and b)
average distance between individuals as a function of generation
- 104 -
The percentage of generation which met the design criteria (10.7% of 100%) is shown
in Figure 6.10. The optimal design values for five suspension design parameters are
consistently near their lower limit, except for the sprung mass which is near to the upper limit.
These optimized design variables provide a “soft suspension performance” regarding driving
safety, thus ensuring suitable forces to keep the tire in contact with the ground’s surface,
whilst on other hand providing better driving comfort.
Time ∝
Generation 10.7%
0 20 40 60 80 100
% of criteria met
Figure 6.10: Percentage of generation as a function of criteria met over the fourth runs
Figure 6.11 introduces the initial and optimal design variables substituted into
objective functions regarding the standard deviation of the vehicle’s body acceleration - σ ,
and dynamic tire forces - as a function of the vehicle velocity. It can be observed that the
reduction of the vehicle’s body acceleration is 50.79%, whilst for the dynamic tire forces it is
53.01%, compared to the initial design of the passive linear suspension system, respectively.
75.0 24000 Initial

Vehicle’s body acceler. [ms-2]
Dynamic tire forces [N]
Optimal
62.5 20000
50.0 16000
37.5 12000
25.0 8000
12.5 4000
0.0 0.0
0 30 60 90 120 150 180 0 30 60 90 120 150 180
-1 -1
a) Vehicle speed [kmh ] b) Vehicle speed [kmh ]
Figure 6.11: Initial and optimal values for the objective functions, as functions of vehicle
speed; a) Vehicle’s body acceleration - , and b) Dynamic tire forces -
- 105 -
6.4.2 Optimization of the design parameters for the active and semi-active non-linear
suspension systems
The stochastic parametric optimization method is applied to ensure the optimal active
damping force. This method is based on non-linear programming, by using the Hooke-Jeeves
method. The optimization procedures are carried out in this order:
- definition of the input data and limit values of the design variables,
- definition of the objective function,
- simulation of the ground excitation, and
- simulations of the different equations of motion.
A flowchart diagram of the optimization process is shown in Figure 6.12. Simulations
of the differential equations of motion are performed by equations (6.17) and (6.18), and the
vehicle’s suspension parameters tabulated in Table 6.4, Pajaziti [23], Lajqi and Pehan [2].
START
Input Ground
data excitation
Hooke-Jeeves
optimization method
Simulation of differ.
equations of motion
Objective Objective function
function = 1020
NO Variable
constraint
YES
Stopping NO
criteria’s
YES
Optimal
parameters
END
Figure 6.12: Flowchart diagram for optimization design variables for damping force
The objective function that enables minimization of the vertical vehicle’s acceleration
regarding the sprung and un-sprung masses is written as follows:
min f ( x ) = &z&2s +α 1 ⋅&z&u2 , (6.29)
where α1 =5 is the weight coefficient that describes the influence ranking of the corresponding
ground excitation (equation (6.23)) on the objective function “f(x)”.
- 106 -
Table 6.4: Suspension parameters used for optimization of the active damping-force
Suspension parameters Symbol Unit Value

Sprung mass ms [kg] 295
Un-sprung mass mu [kg] 35
Shock absorber non-linear damping coefficient csh1 [Nsm-1] 3482
Shock absorber non-linear square damping coefficient csh2 [Ns2m-2] 580
Spring stiffness non-linear coefficient ks1 [Nm-1] 15302
Spring stiffness non-linear square coefficient ks2 [Nm-2] 2728
Tire stiffness non-linear coefficient kt1 [Nm-1] 60063
Tire stiffness non-linear square coefficient kt2 [Nm-2] 42509
Tire stiffness non-linear cube coefficient kt3 [Nm-3] 22875
The design variables’ limits are, Pajaziti [67], Lajqi and Pehan [72]:
0.0 Ns / m ≤ bs ≤ 5000 Ns / m ,
(6.30)
0.0 Ns / m ≤ bu ≤ 2800 Ns / m .
The optimization procedure starts by employing three initial values of the design
variables. The optimal values are reached when the difference between the two adjacent
values of the objective function reaches 10-9. Table 6.5 gives the initial and optimal design
variables. The best results for active damping force Fa regarding driving comfort and driving
safety were achieved by the 2nd iteration process and are bs = 841 Nsm-1 and bu = 2512 Nsm-1.
Table 6.5: Initial and optimal design variables for the active damping force
Parameters 1st option 2nd option 3rd option

Initial design variables:
bs* [Nsm-1] 0.00 2500 5000
bu* [Nsm-1] 0.00 1400 2800
Optimal design variables:
bs [Nsm-1] 1895 841 1198
bu [Nsm-1] 2181 2512 2607
Objective function: f(x) 5.162995814⋅E-01 4.5684816⋅E-01 5.04968724⋅E-01
Number of iteration: 232 221 201
6.5 Simulation of the vehicle’s suspension by using MATLAB/Simulink
The flowchart diagram for simulations of the terrain vehicle’s suspension behavior is
performed in MATLAB/Simulink environments, Figure 6.13 and Figure 6.14. Figure 6.13
shows a flowchart diagram for the initial and optimal design parameters for the passive linear
suspension system. Figure 6.14 present a flowchart diagram for passive, active, and semi-
active non-linear suspension systems.
- 107 -
Figure 6.13: Flowchart diagram for simulation ground excitation, the vehicle’s body
accelerations, velocities, displacements, and tire forces as functions of time for initial and
optimal design parameters of the passive linear suspension system
Figure 6.14: Flowchart diagram for simulation ground excitation, vehicle’s body
accelerations, velocities, displacements, and vertical tire forces as functions of time
for non-linear active, semi-active, and passive suspension systems
These flowchart diagrams simulate the vehicle’s accelerations, velocities,
displacements, and forces powered by ground excitation, depending on time. The excitation is
represented by repeated cosine road bumps. The driving comfort is represented by the
vehicle’s body acceleration (Figures 6.16 and 6.20), Mastinu et al. [46], Lajqi et al. [75],
Popp and Schiehlen [76], where the greater values are undesirable. The driving safety relates
to the vertical tire forces acting between the ground surface and the tire (Figures 6.19 and
Figures 6.23), where it is necessary to be as stable as possible, Belingardi and Demic [69]. If
- 108 -
the vertical tire forces oscillate too much then the contact between the tire and ground surface
will be weak. The vertical tire forces are strictly related to active safety, Mastinu et al. [46].
The linear Fz-l (t) and non-linear Fz-n(t) vertical tire forces are composed from static Fsta
and linear Fdyn-l(t), and non-linear Fdyn-n(t) dynamic ones, given by the following expression:
Fz−l (t) = Fsta + Fdyn−l (t) = g ⋅ (ms + mu ) + kt ⋅[zu (t) − zr (t)],

Fz−n (t) = Fsta + Fdyn−n (t) = g ⋅ (ms + mu ) + kt1 ⋅ [zu (t) − zr (t)]+ kt 2 ⋅ [zu (t) − zr (t)] 2 K (6.31)
3
− kt 3 ⋅ [zu (t) − zr (t)] .
In order to describe the vehicle’s behavior as it runs over road bumps, the ground
excitation is simulated by equation (6.22) and the data given in Table 6.1. The vehicle passed
over the first bump at a speed of 20 kmh-1 and the second bumps at a lower speed of 10 kmh-1.
Figure 6.15 shows the ground excitation derived from a cosine function with amplitudes
H1 = 0.05 m and H2 = 0.1 m for two different vehicle speeds (v1 = 20 and v2 = 10 kmh-1).
0.125
v1 = 20 [kmh-1]
v2 = 10 [kmh-1]
Ground excitation [m]
0.100
0.075
0.050
0.025
0.0
-0.025
0 1 2 3 4
Time [s]
Figure 6.15: Ground excitation as a function of time
Figures 6.16, 6.17 and 6.18 shows the accelerations, velocities, and displacements of the
vehicle’s body, depending on time caused from double cosine road bumps for initial and
optimal design parameters of the passive linear suspension systems.
27.0
Vehicle body acceler. [ms-2]
Initial Initial
18.0 22.22 ms-2 Optimal
Optimal -75%
9.0 5.56 ms-2
0.0
-9.0
-18.0
-27.0
0 1 2 3 4
Time [s]
Figure 6.16: Vehicle’s body accelerations as a function of time for passive linear system
- 109 -
1.2
Initial
Vehicle body velocity [ms-1]

0.8 Optimal
0.4
0.0
-0.4
-0.8
-1.2
0 1 2 3 4
Time [s]
Figure 6.17: Vehicle’s body velocities as a function of time for passive linear system
0.13
Initial
0.10 Optimal
0.07
0.04
0.01
-0.02
-0.05
0 1 2 3 4
Time [s]
Figure 6.18: Vehicle’s body displacements as a function of time for passive linear system
Figure 6.19 shows the vertical tire forces depending on time, for the initial and optimal
design parameters for passive linear suspension systems.
1250
Initial
Optimal
9500
Vertical tire forces [N]
6000
2500
-1000
-4500
0 1 2 3 4
Time [s]
Figure 6.19: Vertical tire forces as a function of time for passive linear system
- 110 -
Figures 6.20, 6.21 and 6.22 show the accelerations, velocities, and displacements of the
vehicle’s body, depending on time, caused from double cosine road bumps for passive, active,
and semi-active non-linear suspension systems.
8.0
Vehicle body acceleration [ms-2]

Passive
Semi-active Semi-active
5.0 -31% Active
2.0
-1.0
-4.0 Active
-54%
-7.0
-10.0
0 1 2 3 4
Time [s]
Figure 6.20: Vehicle’s body accelerations as a function of time
1.0
Passive
Vehicle body velocity [ms-1]
Semi-active
0.6
Active
0.2
-0.2
-0.6
-1.0
0 1 2 3 4
Time [s]
Figure 6.21: Vehicle’s body velocities as a function of time
0.13
Passive
0.10 Semi-active
Active
0.07
0.04
0.01
-0.02
0.05
0 1 2 3 4
Time [s]
Figure 6.22: Vehicle’s body displacements as a function of time
- 111 -
7000
Passive
6000
Vertical tire forces [N]

Semi-active
Active
5000
4000
3000
2000
1000
0
0 1 2 3 4
Time [s]
Figure 6.23: Vertical tire forces as a function of time
Figure 6.23 shows the vertical tire forces depending on time, for passive, active, and
semi-active non-linear suspension systems.
6.6 Discussion of the results
Evaluation of a terrain vehicle's suspension system was performed by simulation of the

double cosine road bumps. The vehicle passed over the first bump at an amplitude of 0.05 m
and a speed of 20 kmh-1, and the second bump at an amplitude of 0.1 m and a speed of 10
kmh-1, Figure 6.15. This simplified form of the repeated road bumps with different amplitudes
is used for better understanding of the appearances of the vehicle’s body accelerations,
velocities, displacements, and vertical tire forces for active, semi-active, and passive systems.
Figures 6.16, 6.17 and 6.18 introduce the accelerations, velocities, and displacements of
the vehicle’s body, as a function of time, for initial and optimal design parameters for passive
linear suspension system. The references are for the initial design. When the vehicle passes
over the first bump, the values for the vehicle’s body: accelerations are reduced 37%, 45%,
and 75%, velocities 28%, 45%, and 68%, and displacements 15%, 32% and 49% (depending
on the location). The same explanation can also be found when the terrain vehicle passes over
the second bump due to lower speed and higher amplitude. Figure 6.19 shows the vertical tire
forces as a function of time, for initial and optimal designs of the passive linear suspension
system. If the vehicle passes over the first bump, the vertical tire forces at maximal peaks is
reduced by 84%, whilst at the second bump by 71% when is compared to the initial ones.
Figure 6.20 shows the vehicle’s body accelerations for passive, active, and semi-active
non-linear systems. The reference acceleration is the passive one. If the vehicle passes over
the first bump the active system reduces it by 40%, 54%, and 81% (depending on the
- 112 -
location). The semi-active system reduces it by 31%, 46%, and 76% (depending on the
location). Similar appearances were observed at the second bump, but the speed was lower
and the amplitude higher. The same explanation for the second bump can also be found in
Figures 6.21, 6.22 and 6.23. Figure 6.21 shows the vehicle’s body velocities for passive,
active, and semi-active non-linear systems. The reference velocity is the passive one. When
passing over the first bump, the active suspension system reduced it by 44% and 81%. The
semi-active system reduced it by 36% and 71%. Figure 6.22 shows the vehicle’s body
displacements for passive, active, and semi-active non-linear systems. The reference
displacement is the passive one. When the vehicle passed over the first bump the active
system reduced it by 43%. The semi-active system reduced it by 38%. Figure 6.23 shows the
vertical tire forces for passive, active, and semi-active non-linear systems. The reference force
is the passive one. When passing over the first bump the active system reduced it by 45% and
54%. The semi-active system reduced it by 22% and 49%.
If the terrain vehicle equipped with active or semi-active systems is driven over the road
bumps at 10 and 20 km/h, the driving comfort ( &z&s < 4 ms-2) and the driving safety
(Fz < ± 1.4 g), according Kuznestov et al. [81] are as yet within the acceptable zone.
Although the passive linear system is optimized by MOGAs and the results are satisfactory
compared with the initial design, the performances offered by this system are lower than the
active or semi-actives suspension systems. If the terrain vehicle is equipped with a passive
linear or non-linear system, the acceleration and dynamic tire forces, according to Kuznestov
et al. [81], has exceeded the zone limit and consequence have negative influences on the
driving comfort and driving safety.
Finally, it can be concluded that, the active system provides better performance in
comparison with the passive system on account of better isolation of the vibration. Due to the
damping capability, the semi-active systems were less effective than the active ones, as can be
seen from the presented diagrams.
- 113 -
7 DEVELOPMENT OF THE FOUR WHEELS STEERING

SYSTEM FOR THE 4X4 TERRAIN VEHICLE
This chapter introduces the development of the four/two wheels steering system intended to
be used for 4x4 terrain vehicle. The four wheels steering (4WS) concept comes into
consideration when the driver’s steering wheel is able to turn both the front and rear wheels of
the vehicle during the same steering input. In most cases the 4WS system operates using
three modes, such as counter steer, neutral steer, and same steer, Figure 7.1.
counter neutral same

steer steer steer
a) b) c)
Figure 7.1: Operation modes of the 4WS; a) counter steer, b) neutral steer, and c) same steer
The counter steer mode is the case when the rear wheels are turned in the opposite
direction to the front wheels. The neutral steer mode is a situation when the rear wheels are
locked and the vehicle operates as an ordinary vehicle. The same steer mode comes into
consideration when the rear wheels are turned in the same way as the front ones. The counter
steer mode of the 4WS system is usually used at low vehicle speed, the neutral steer mode for
average vehicle speed, whereas the same steer mode is for higher vehicle speed, Alter [40].
The main cause that affects the marketing limitations of vehicles with 4WS has been the
vehicle’s instability. On the one hand, the stability of the vehicle is one of the key issues for
driving safety whilst, on the other hand, the requirements given by Council Directive
70/311/ECC and the latest supplement of the Commission Directive 1999/7/ECC regarding
the steering system for vehicle’s [82] has led the manufacturers of vehicles to limit the
applications of 4WS.
The objective of this approach is to develop an effective steering system for 4x4 terrain
vehicle, which offers high rates of the maneuverability when cornering, driving safely, and
fulfilling the directives described previously. Based on the design task and operational
- 114 -
conditions, the steering mechanism for the terrain vehicle will be designed for two modes.
One is to only steer the front wheels - 2WS and the other to steer all four wheels - 4WS,
Figures 7.1.a and 7.1.b. The main target will be to design steering mechanisms that provide
Ackermann steering geometry exactly. In order to actually utilize these types of steering
system within terrain vehicles it is necessary to be clear that these systems do have some
weaknesses especially where there is a question of stability. Main weakness has been shown
to be bump steering as the result of significant vertical wheel motion. On the other hand, a
terrain vehicle requires the vertical motion of its wheels to be very extensive due to vehicle
rollover risk. Is it necessary to design a totally new concept? This approach is necessary in
order to find a unique solution for the steering mechanism that exactly ensures the Ackermann
conditions for the two modes. An adequate solution can be assumed by designing several
pairs of gear transmissions such as bevel, planetary, and non-circular gears. Furthermore,
when developing a 4WS system it must be in relation to the vehicle’s kinematic behavior.
Therefore, the following derives kinematic relations for two/four wheels steering system.
7.1 Kinematics of the two/four wheel steering system
For all vehicles to maneuver using it, it is necessary to have a steering mechanism in order to
turn the steered wheels. The main task of this mechanism is to steer the vehicle and to follow
an advance prescribed trajectory, Pehan et al. [1]. This should be done easily and safety by
steering either the front or both the front and rear wheels. The kinematic behavior of the
steering mechanism systems for 2WS and 4WS are given in the follow for a concrete terrain
vehicle. Performances provided by terrain vehicles with 4WS will be evaluated by
comparisons with the 2WS system.
7.1.1 Kinematic of the two-wheel steering system
The kinematic behavior of the 2WS system is determined when the rear wheels are not steered
and the sideslip angle of the wheels is assumed to be negligible. This comes into
consideration when the vehicle moves slowly, Spentzas et al. [41].
In order for all the wheels to turn freely on a curved road, the normal lines to the centers
of each wheel must intersect at a common point, known as the vehicle turning center O. This
is known as Ackermann steering geometry. Determining the correct steering geometry was a
major problem in the early days the vehicles’ manufacturers. Figure 7.2 illustrates the case
when the terrain vehicle turns towards the left side using the 2WS system. The front steering
angles for the left and right wheels are derived from the triangles ∆OAC and ∆OBD, such as:
- 115 -
2⋅l
tan δ L = , (7.1)
2 ⋅ R2WS − w
2⋅l
tan δ R = , (7.2)
2 ⋅ R2WS + w
where δL and δR denote the front steering angles for the left and right wheels, respectively.
R2WSmax
w
δL δR
A B
R2WS
lF
CG
R2WSmin
l
lR
δL δR
O C D
Figure 7.2: Turning of the terrain vehicle towards the left side using the 2WS system
The turning radiuses for the vehicle with 2WS presented in Figure 7.2 are determined
by the following expressions:
l w l w
R2WS = + = − , (7.3)
tan δ L 2 tan δ R 2
l a+b
R2WS min = = , (7.4)
tan δ L tan δ L
l a+b
R2WS max = = , (7.5)
sin δ R sin δ R
where R2WS is the turning radius at the center of the rear axle, whilst R2WSmin and R2WSmax
denote the minimal and maximal turning radius for the vehicle with 2WS, respectively.
- 116 -
After some mathematical arrangement in equation (7.3) it gets the Ackermann steering
condition between the left and right wheel δL and δR, presented by the following expression:
w (7.6)
cot δ R − cot δ L = .
l
Equation (7.6) presents Ackermann’s steering geometry or Ackermann steering

condition. The Ackermann steering geometry is valid when the vehicle moves slowly and the
sideslip angle is negligible.
7.1.2 Kinematic of the four-wheel steering system
The greater maneuverability of the terrain vehicle is one of the important issues whether it
operates over the roughest terrain or in urban areas. A higher rate of the vehicle
maneuverability is obtained when all the wheels (4WS) turn around their kingpin axes and the
rear wheels are go in opposite directions to the front ones. Figure 7.3 presents situations when
the front and rear wheels of the terrain vehicle go in opposite directions, known as counter
steer mode that produces a higher maneuverability rate.
In order to determine expressions for estimating a vehicle’s turning radius with four
wheels steering (R4WS), it is necessary to start by the assumption that the sideslip angle is zero.
If the vehicle moves up to a 50 km/h or lower speed, the sideslip angle of the wheels may be
negligible, Alter [40], Spentzas et al. [41].
w
δLF δRF
A B
lF
c
δLF δRF
O C CG D
l
δLR δRR R4WS

lR
δLR δRR
d
R4WSmin
E F
R4WSmax
Figure 7.3: Turning of the terrain vehicle towards the left side using the 4WS system
- 117 -
In order to ensure that all the wheels of the vehicle can be turned freely on a curved road
and fulfill the Ackermann steering geometry, the perpendicular lines of the centers of each
wheel must intersect at a common point. The intersection point O is known as the vehicle
turning center. The front steering angles for the left and right wheels δLF and δRF are derived
from the triangles ∆OAC & ∆OBD. The rear steering angles for the left and right wheels δLR
and δRR are determined from triangles ∆OCE & ∆ODF. The expressions for determining the
steering angles of the wheels are shown in Figure 7.3, and written as follow:
2⋅c
tan δ LF = , (7.7)
2 ⋅ R4WS − w
2⋅c
tan δ RF = , (7.8)
2 ⋅ R4WS + w
2⋅d
tan δ LR = , (7.9)
2 ⋅ R4WS − w
2⋅d
tan δ RR = , (7.10)
2 ⋅ R4WS + w
where δLF and δRF denote the front steering angles for the left and right wheels, and δLR and
δRR the termed rear steering angles for the left and right wheels, respectively. The parameters
c and d indicate the longitudinal distances between the turning centers of the front and rear
wheels (point A or B and E or F) and the vehicle turning center O.
The turning radiuses for the vehicle with the 4WS system shown in Figure 7.3 are
written by the following expressions:
c w c w
R4WS = + = − , or (7.11)
tan δ LF 2 tan δ RF 2
d w d w
R4WS = + = − , (7.12)
tan δ LR 2 tan δ RR 2
c d
R4WS min = = , (7.13)
sin δ LF sin δ LR
c d
R4WS max = = , (7.14)
sin δ RF sin δ RR
where R4WS is the turning radius at the center of gravity of the vehicle, whilst R4WSmin&R4WSmax
denote the minimal and maximal turning radius for the vehicle with 4WS, respectively.
After several mathematical operations in equations (7.11) and (7.12), the kinematic condition
between the left and right wheels on the front and rear axles was arrived at, written as follows:
- 118 -
w
cot δ RF − cot δ LF = , (7.15)
c
w
cot δ RR − cot δ LR = . (7.16)
d
Figure 7.3 indicates that, the longitudinal distances c + d is given as follow:
c + d = l. (7.17)
By combining equations (7.15) and (7.16), and substituting equation (7.17),the

following expression is obtained:
(cot δ RR − cot δ LR ) + (cot δ RF − cot δ LF ) = l . (7.18)

(cot δ RF − cot δ LF ) ⋅ (cot δ RR − cot δ LR ) w
Equations (7.18) present the kinematic relation between the steering angles of the front
and rear wheels for a counter steer mode.
7.2 Steering system for a terrain vehicle
Terrain vehicle is designed to operate on flat roads as well as rough terrains, on the one hand,
whilst on the other hand the main characteristics of terrain vehicles is their very extensive
wheel movements due to the risk from rollover. For that reason it is required to have robust
and effective steering system which provides as much as possible, maneuverability and
suitable stability. In order to fulfill the predefined requirements, the steering system for a
terrain vehicle is designed to work with two modes. The first mode is the 2WS introduced in
Figure 7.2, whereas the second mode is 4WS, as represented in Figure 7.3.
7.2.1 Determining the steering wheel angles for the terrain vehicle in the 2WS mode
In order to fulfill the requirements given by the check list in Table 1.1, it is necessary to
determine the maximal turning steering angles of the left and right wheels. These steering
angles are determined by using equations (7.1) and (7.2), where it is written in matrix form:
 2⋅l 
a tan
−1
 δ L   1 0   2 ⋅ R2 ws − w 

 =  ⋅ 2⋅l . (7.19)
 δ R   0 1   a tan 
 2 ⋅ R2 ws + w 
By substituting the geometric dimensions given in Table 1.1 into the equations (7.3) to
(7.5) and (7.19) the results for the geometric characteristic of the steering system with 2WS
are obtained, as shown in Table 7.1.
- 119 -
Table 7.1: The geometric characteristics of the steering system with 2WS
Turning radius for 2WS, [m] Steering angles for 2WS, [0]
Parameters →
R2WSmin R2WS R2WSmax δLmax δRmin δLmin δRmax
Values → 2.5 3.55 5.39 -48.24 -31.33 31.33 48.24
Figure 7.4.a presents the case where the front wheel of the vehicle is turns towards the
left side, and in Figure 7.4.b case when it turns towards the right side. The maximal turning
radius on the left or right side from a straight-ahead position is ensured when the driver turns
the steering wheel by φmax = ±5400 or 1.5 [rot] towards the left or right, Figure 7.4.
δLmax a) b) δRmax
δRmin δLmin
iLeftmin iRightmax iLeftmax iRightmin

(-) (+)
Left steering Right steering

wheel wheel
φLmax = -5400 Steering wheel φRmax = 5400
Figure 7.4: Turning of the vehicle with 2WS; a) on the left side, and b) on the right side
The conventional sign used for steering angles (δL and δR) will be negative (-) if the
front wheels turn towards the left side, otherwise positive (+). The maximal steering angles of
the left wheel when the vehicle is turning towards the left side is δLmax = -48.240, whilst the
right wheel has minimal values δRmin = -31.330. Similar appearances were ensured when the
vehicle turns towards the right side and the steering angles are δLmin = 31.330 & δRmax = 48.240.
According of the Ackermann steering geometry, the steering gear ratios are different for
the left and right wheels. The steering gear ratio i is the steering wheel angle ϕ divided by the
front steering angles of the wheels, written as follow:
ϕ L max ϕ R max
iLeft min = iRight min = = = 11.20 , (7.20)
δ L max δ R max
ϕ R max ϕ L max
i Left max = i Right max = = = 17.24 . (7.21)
δ L min δ R min
Equations (7.20) and (7.21) show that the steering mechanism, in order to fulfill the
Ackermann steering conditions, should be worked out as multiplication/reduction.
- 120 -
Figure 7.5 illustrates the Ackermann steering condition for the vehicle with 2WS, as
computed by equation (7.6). The obtained results for the front steering angle of the right
wheel δR when the vehicle turns towards the left or right sides are in dependence from the
steering angle of the left wheel δL and on the function of the steering wheel angle φ. Figure
7.5 shows that the front steering angle of the right wheel δR displays non-linear behavior
compared with the input linear steering angle of the left wheel δL.
50
δLmax δRmin δL
(-) δR
40
Steering angle of the left & right wheel δL & δR [0]
Turn toward  w
δ R (δ L ) = a cot  cot δ L − 
the left side  l 
30
20
δL = -48.24...00
10
φ = -540...00 δL=0...31.330
-540 -480 -420 -360 -300 -240 -180 -120 -60 0 60 120 180 240 300 360 420 480 540
 w δLmin δRmax
δ R (δ L ) = a cot  cot δ L +  -10
 l  (+)
Turn toward
-20
the right side
-30
δL = -48.24...00 δL=0...31.330
-40
φ = 0...5400
-50
Steering wheel angle φ [0]
Figure 7.5: Steering angles for the front wheels as a function of the steering wheel angle φ
7.2.2 Determining of the steering wheel angles for the terrain vehicle in the 4WS mode
The maximal turning steering angles for the left and right wheels on the front and rear axles
(δLF, δRF, δLR and δRR), as shown in Figure 7.3, are determined by employing equations (7.7)
to (7.10), where in matrix form they are written as follows:
−1
 δ LF   1 0 0 0   a tan( 2⋅c 
     ) 
2 ⋅ R4 ws − w
 δ RF   0   
  1 0 0  2⋅c 
a tan( )
     2 ⋅ R4 ws + w  (7.22)
 =  ⋅ .
 δ LR   0 0 1 0   a tan(
2⋅d
) 
     2 ⋅ R4 ws − w 
     2⋅d 
 δ RR   0 0 0 1   a tan( 2 ⋅ R + w ) 
     4 ws 
- 121 -
The decisional characteristics of the steering system with 4WS are determined by the
substituting the geometric dimensions given in Table 1.1 into equations (7.11) to (7.14) and
(7.22). The main geometric characteristic of the 4WS system is presented in Table 7.2.
Table 7.2: The main geometric characteristics of the steering system with 4WS
Turning radius of 4WS, [m] Steering angles for 4WS, [0]

Parameters →
R4WSmin R4WS R4WSmax δLF δRF δLR δRR
L -34 -18.5 34 18.5
Values → 2.5 3.13 4.4
R 18.5 34 -18.5 -34
Figure 7.6.a introduces steering angles of the wheels when vehicle is turning towards
the left side, while Figure 7.6.b when vehicle is turning towards the right side.
δLFmax a) δRFmin δLFmin b) δRFmax
(-) (+)
φLmax = -5400 Steering wheel φRmax = 5400
(+) (-)
δLRmin δRRmax
δLRmax δRRmin
Figure 7.6: Turning of the 4WS system; a) towards the left side, and b) towards the right side
When the vehicle is turning towards the left side, the maximal steering angles of the left
and right wheels for the front axle are δLFmax = -340 and δRFmin = -18.50, whilst for the rear axle
they are δLRmax = 340 and δLR min = 18.50. Similar appearances are ensured when the vehicle
turns towards the right side and the steering angles are δLFmin = 18.50, δRFmax = 340,
0 0
δLRmin = -18.5 and δLRmax = -34 , Figure 7.6. The steering gear ratios i for the vehicle with
4WS are determined by the following expressions:
ϕ L max ϕ R max ϕ L max ϕ R max

iF min = iR min = = = = = 15.88, (7.23)
δ LF max δ RF max δ LR max δ RR max
- 122 -
ϕ R max ϕ L max ϕ R max ϕ L max

iF max = iR max = = = = = 29.19. (7.24)
δ LF min δ RF min δ LRmin δ RR min
From equations (7.23) and (7.24) it is shown that, the steering mechanism in order to
fulfill Ackermann steering condition should be worked out as multiplication/reduction.
Figure 7.7 illustrates the Ackermann steering condition for the vehicle with 4WS
computed by equations (7.15) and (7.16). The obtained results of the steering angles for the
front and rear right wheels δRF and δRR when the vehicle turns towards the left or right sides
are dependent on the input steering angles of the front and rear left wheels δLF and δFR as a
function of the steering wheel angle φ. From Figure 7.7, it can be observed that the steering
angles for right wheels of the front and rear axles have non-linear behavior compared with the
input linear steering angles for the left wheels on the front and rear axles.
35 35
δLF δLR=0...340 Rear wheels δLR
Steering angles δLR &δRR [0]
Steering angles δLF &δRF [0]
28 28 turns left side δRR

Front wheels δRF (-)
turns left side 21 δ 21
(-) 14 14
7 7
δLF=0...18.50
-540 -450 -360-270-180 -90 0 90 180 270 360 450 540 -540 -450 -360-270-180 -90 0 90 180 270 360 450 540
-7 -7
-14 -14
-21 (+) -21 δLR=-18.5...00
(+)
-28 Front wheels Rear wheels -28
δLF=-34...00 turns right side turns right side
-35 -35
a) Steering wheel angle φ [0] b) Steering wheel angle φ [0]
Figure 7.7: Steering angles for the front and rear wheels as a function of the steering wheel
angle φ when vehicle with 4WS turns; a) towards the left side, and b) towards the right side
Figure 7.8 presents the cases when the terrain vehicle turns towards the left side and the
steering systems operate as 2WS and 4WS modes. The vehicle turning radius is minimal
when the steering wheel is rotated by φmax = - 5400 and the vehicle provides higher
maneuverability. When the terrain vehicle operates in 4WS mode, the turning radius is
reduced by 34% compared with 2WS mode. This higher maneuverability only comes into
consideration when the vehicle is moving at a slower speed.
Figure 7.9 shows the maximal vehicle turning radius R2WSmax and R4WSmax for both
modes 2WS and 4WS as a function of the steering wheel angle φ. From the curves it can be
observed that the vehicle’s turning radius has smaller values when all the wheels are steered
(4WS) thus ensuring a higher rate of maneuverability by the terrain vehicle. When the vehicle
moves at a higher speed, the rear steered wheels are locked and steering system is in 2WS
mode in order for the vehicle to provide better stability due to the increasing turning radius.
- 123 -
1.90 m
2.89 m
R2WSmax R2WSmin
2.5 m
2.5 m
340
18.50
2.8 m
4WS
0 O 18.50
31.33
340
φmax = 5400
2WS 48.240
R4WSmin
R4WSmax
2.1 m
Figure 7.8: Terrain vehicle’s turning radius when operates in the 2WS and 4WS modes
90
R2SWmax – Turning radius for 2WS
Vehicle turning radius R2SWmax or R4SWmax [m]
R4SWmax – Turning radius for 4WS

75
l a+b
60 R2WS max = =
sinδ R sinδ R
c d
45 R4WSmax = =
sinδ RF sinδRR
30
15
0
0 60 120 180 240 300 360 420 480 540
0
Steering wheel angles, φ [ ]
Figure 7.9: Vehicle turning radius as a function of steering wheel angle φ for 2WS and 4WS
- 124 -
7.3 Design of the steering mechanism for a terrain vehicle
There are devised many different designs for the steering mechanism. Even though they have
the same basic components, each vehicle has some requirements that differ from others. For
this reason a steering mechanism must be modeled and designed for each vehicle.
Each type vehicle has slightly different dimensions (length, width, etc.), different
suspension systems and so on, which make it necessary to design new steering mechanisms.
Even small changes in the dimension of the vehicles can drastically alter the steering
performance, Leishman [83].
There are several types of steering mechanism used on road vehicles such as rack and
pinion, lever arm, drag link, and the multi-link, Rill [4] and Jazar [9], Figure 7.10. Each of
them has some advantages and disadvantages. Obviously the rack and pinion steering
mechanism is the more widely used as a steering system within passenger vehicles, Figure
7.10.a. It is simple in design, has higher mechanical efficiency and lower price compared with
other systems. The lever arm steering mechanism is used on tracks with large wheelbases and
an independent suspension system at the front axle, Figure 7.10.b. The drag link steering
mechanisms are usually used for trucks and buses with a front solid axle. By using such a
steering mechanism the steering command is sent to only one wheel and connects the other
one by a drag link, as shown in Figure 7.10.c.
a) b) Steering gearbox
Drag link uR φs
uR Steering lever
Steering
δL Rack gearbox Pinion δR δL φLev δR
Drag link
Kingpin axis Kingpin axis
c) d) Steering lever
φLev Steering gearbox φLev
Steering link
Steering lever
Drag link
Steering link
δL δR δL δR
uR
Drag link
Figure 7.10: Road vehicle steering mechanism; a) rack and pinion, b) lever arm,
c) drag link and d) multi-link
- 125 -
The multi-link steering mechanisms are used for busses and big trucks where it is
necessary to have large steering angles at the front wheels in order to achieve higher
maneuvering, Figure 7.10.d. Practically, it is impossible to build a simple steering mechanism
in order to provide exactly the Ackermann steering condition. However, various steering
mechanisms are used in practice, more or less satisfying the Ackermann steering conditions.
Let us model one of the better known steering mechanisms which are used in almost all types
of road vehicles, known as the trapezoidal steering mechanism.
7.3.1 Design of the trapezoid steering mechanism for front wheels steering
The Ackermann steering condition could be obtained if the trapezoidal steering mechanism
were to be optimized. This mechanism would work closely to the Ackermann condition and
would be exactly at a few angles, Figure 7.11. However, a trapezoidal mechanism is simple
enough to produces in mass, and provides suitable accuracy for use in vehicles.
a)
w - 2⋅r⋅sinβ r Axle bearing
Kingpin axis β β
Drag link
w
r δR
b)
w = 2100 [mm]
δL r l = 2800 [mm]
ξ
β δL = - 48.24...31.33 [0]
δL δR r = 415 [mm]
β β = 18, 22, 26, 30 [0]
Figure 7.11: Trapezoidal steering mechanism;

a) straight-ahead position, b) turning toward the left side
Figure 7.11.a illustrates a trapezoidal steering mechanism in the straight-ahead position.

Two characteristic parameters exist such as: angle β and the length of the offset arms r. Figure
7.11.b shows the steerable position as the trapezoidal mechanism turns towards the left side
by the illustration of steer angles of the left and right wheels δL and δR and other parameters.
In order to obtain a relation between the left and right steering angles (δL+β) and
(δR-β) as shown in Figure 7.11.b, the projections of the Cartesian coordinative system in x and
y axes are written as follows:
- 126 -
x : r ⋅ cos (δ L + β ) + ( w − 2 ⋅ r ⋅ sin β ) ⋅ cosξ − r ⋅ cos (δ R − β ) = 0,

(7.25)
y : r ⋅ sin (δ L + β ) + ( w − 2 ⋅ r ⋅ sin β ) ⋅ sinξ − r ⋅ sin (δ R − β ) = w.
After some mathematical operation, the relationship between the left and right steer
angles (δL+β) and (δR-β) now has got this form:
r r r
sin (δ L + β ) − sin (δ R − β ) + ⋅ cos (δ R − δ L − 2 ⋅ β ) = + 2 ⋅ sin β − 2 ⋅ + 2 ⋅ sin 2β . (7.26)
w w w
Equation (7.26) is known as the Freudenstein's Equation. The functionality of the

trapezoid steering mechanism for a concrete terrain vehicle compared to the Ackermann
steering condition is presented in Figure 7.12. It depicts that for the given parameters shown
in Figure 7.11.b, a trapezoidal steering mechanism with β ≈ 220 provides approximately the
best Ackermann condition for the same input steer angles of the left wheel δL, Figure 7.12.
35
Steering angles of the right wheel, δR [0]
30
25
20 β = 180
β = 220
15 Ackermann
β = 260
10 β = 300
5
0
0 8 16 24 32 40 48
Steering angles of the left wheel, δL
[0 ]
Figure 7.12: Comparison between functionality of the trapezoidal steering mechanism for
different β angles and Ackermann steering condition
It is necessary to define in order to examine the performance that provides trapezoidal
steering mechanism and compare it with the Ackermann steering condition as error
parameters e. This error is the absolute difference between the steering angles of the right
wheel δR defined by the trapezoidal steering mechanism and the Ackermann steering
condition for the same input steering angles of the left wheel δL.
The error parameter e is determined by the following expression:
e = ∆δR = δR, trapezoidal − δR,Ackerman (7.27)
Figure 7.13 presents the values of the error parameter e steering mechanism that could
be used for a terrain vehicle by using different β angles as a parameter. For different angles β
- 127 -
of the trapezoidal steering mechanism presented by the diagrams in Figures 7.12 and 7.13, it
can be observed that for β = 220 and smaller input of steering angles δL = 0 … 100, the error is
minor. For δL = 280 error has got maximal value e = 1.30, whilst for δL ≈ 440 the error is zero.
7.5
β= 180
β = 220
Ackermann
Error values, e = ∆δR [0]

6.0
β= 260
β = 300
4.5
3.0
1.5
0.0
0 8 16 24 32 40 48
Steering angles of the left wheel, δL
0
Figure 7.13: The error parameter e for concrete trapezoidal steering mechanism
If the terrain vehicle operates on the flat road by using such a mechanism, the vehicle
will ensure sufficient Ackermann condition and at the same time provide suitable stability.
The situation is totally different when the terrain vehicle operates over rough terrain. In this
case, their wheels move extremely in a vertical direction in order for the vehicle to provide
good performance and to be protected from rollover risk. On the other hand, the greater
vertical motion of the wheels will cause so-called bump steering.
Bump steering will be appeared if during the vertical motion of the wheel, steerable
wheels will be rotated for minor angles around the kingpin axle without the driver’s steering
input. In this case the vehicle moves along a different path which has not been selected by the
driver and causes a reduction of vehicle stability, Figure 7.14. Bump steering is an undesired
phenomenon and depends on the design of the suspension and steering mechanisms. To avoid
this phenomenon requires the suspension and steering mechanisms to be rotated at the same
centers during suspension travel. However, it is not that easy to design mechanisms which
avoid bump steering, especially when the vertical movements of the wheels are greater?
Perhaps as a proper solution to avoid bump steering would be the designing of the new
steering mechanism by utilizing some kind of gear transmission. In this situation the steered
wheels during vertical motion (bound and rebound) will stay in the straight-ahead position
without minor rotation around the kingpin axis, such as presented in Figure 7.14.b.
- 128 -
FRONT VIEW
BOUND
REBOUND
Uper control Kingpin axis
arm (Steering axis)
Vehicle
chassis
Steering Lower
knuckle
control arm
Road Flat Road
bump road hole
TOP VIEW
Steering links Wheel centre line
Driving Driving
direction direction
a) e = -∆δ Vehicle b) e=0 c) e = +∆δ

Toe-out centre line Straight-ahead Toe-in
Figure 7.14: Cases of the bump steering; a) toe-out, b) straight-ahead direction and b) toe-in
Toe-out and Toe-in are undesired angles formed between the center line of the wheel
and vehicle in longitudinal direction when the vehicle is viewed from above, Figure 7.14.
7.3.2 Design of a vehicle steering mechanism with 2WS by using non-circular gears
Design of the steering mechanism for terrain vehicle with 2WS is developed in such a way as
to provide exactly the Ackermann steering conditions. By fulfilling this condition this ensures
a reduction of the tire wear and improves the vehicle stability and maneuverability. On the
other hand, vehicle stability would be improved if the steering bump is avoided during greater
vertical motion of the steerable wheels.
However, the commonly used trapezoidal steering mechanism cannot exactly satisfy the
Ackermann steering condition by avoiding steering bump. A perfect vehicle steering
mechanism would be considered if the vehicle followed, as much as possible, the prescribed
path determined by the driver and was unaffected steering bump when the vehicle moves over
different terrains. The proper solution would be achieved by using pairs of non-circular gears.
These pairs have non-circular pitch curves L1 and L2 which ensure variable gear ratios, Figure
7.15. Suitable design of the pairs of non-circular gears would fulfill exactly the Ackermann
steering geometry. The non-circular gears are represented by two roll lines L1 and L2,
corresponding to pitch curves with two fixed rotation centers O1 and O2, rolling together
without slip. In general, the pitch curves L1 and L2 of the non-circular gears are described by
using polar coordinates (r, δ), Figure 7.15.b.
- 129 -
a) b)
2
O2
ω2
L2 r2(δ1)
δ2(δ1) x2
L2
y1 y2
L1 I
a12
Pitch
r1(δ1) δ1
L1 x1
1
ω1
O1
Figure 7.15: a) The non-circular gears, and b) pitch curves of the non-circular gears
The first pair of the non-circular gears is designed to rotate left front wheel, whilst the
other pair to rotate the right front wheel, Figure 7.16. The angular displacements of the pair of
non-circular gears δ1 and δ2 are written by the following expression:
δ1 = δ R ,
(7.28)
δ2 = δ L ,
where δ1 and δ2 represents the angular displacements of the driving and driven non-circular
gears 1 and 2, whilst δR and δL define front steering angles of the left and right wheels.
The angular displacement of the driving gear δ1 is the input value. Table 7.1 presents
the calculation values for the front steering angles that are substituted by angular displacement
of the non-circular driving and driven gears δ1 = δR and δ2 = δL. For the case when the angular
displacement δ2 = -δL is counterclockwise then vehicle turns towards the left side, Figure
7.16.a. If the angular displacements are equal to zero then the vehicle goes in a straight-ahead
direction, Figure 7.16.b. Whilst for the case when the angular displacement δ2 = δL is
clockwise, the vehicle turns towards the right side, Figure 7.16.c.
The angular displacement of the non-circular driven gear δ2 as a function of the driving
gear δ1 derived from Ackermann steering geometry, is determined as follow:
a cot (cotδ1 − w / l ) if − δ1min ≤ δ1 ≤ 0 ,

δ 2 (δ1 ) =  (7.29)
a cot (cotδ1 + w / l ) if 0 ≤ δ1 ≤ δ1max .
- 130 -
δ2=δLmax δ2=δRmin
a)
O2L O2R
δLmax δRmin
(-)
δ1 δ1
Left front
O1L O1R
steering wheel
δ2=00 δ2=00
b)
O2L O2R
δL=00 δR=00
δ1=00 δ1=00
O1L O1R
δ2=δLmin δ2=δRmax
c)
O2L O2R
δLmin δRmax
(+)
δ1 δ1
O1L Right front O1R
steering wheel
Figure 7.16: Working position of the two pair’s non-circular gears when the front wheels of
the terrain vehicle are turning towards the: a) left side, b) straight-ahead, and c) right side
The gear’s ratio i12(δ1) for the driving gear to the driven gear of the non-circular gears is
ω1 (δ1 ) dδ1 / dt dδ 1 r2 (δ 2 (δ1 ))

i12 (δ1 ) = = = = , (7.30)
ω2 (δ 2 ) dδ 2 / dt dδ 2 r1 (δ1 )
where ω1(δ1) and ω2(δ2) denotes angular velocity, whilst r1(δ1) and r2(δ2(δ1)) represents the
radius of the pitch curves for the driving and driven non-circular gears, respectively.
- 131 -
The center distance a12 between the two fixed rotational centers O1 and O2 of the non-
circular gears (Figure 7.15.b) always remains at a constant value and is determined as follow:
r1(δ1 ) + r2 (δ2 (δ1 )) = a12 = const, (7.31)
where a12 is distance between the rotation centers of the non-circular gears.
The radius of the pitch curves for driving gear r1(δ1) and driven gear r2(δ2(δ1)) (Figures
7.15.b) are derived from equations (7.30) and (7.31), written as follow:
a12 a12
r1 (δ1 ) = = ,
1 + i12 (δ1 ) 1 + 1/(dδ 2 / dδ1 )
(7.32)
i12 (δ1 ) ⋅ a12 a12 /(dδ 2 / dδ1 )
r2 (δ 2 (δ1 )) = = ,
1 + i12 (δ1 ) 1 + 1/(dδ 2 / dδ1 )
where the derivative function dδ2/dδ1 is defined from equation (7.29), and is written using the
dδ 2 (δ1 ) 1/ sin δ1 ⋅ 1 + (cotδ1 − w / l ) if − δ1min ≤ δ1 ≤ 0,

=
2
[ 2
] (7.33)
dδ1 1/ sin2 δ1 ⋅ 1 + (cotδ1 + w / l )2 if 0 ≤ δ1 ≤ δ1max . [ ]
For modeling of the non-circular gear in Table 7.3, the necessary parameters have to be given.
Table 7.3: Parameters for modeling non-circular gears for steering system with 2WS
Parameters for modeling non-circular gears

Parameters → 0
w [mm] l [mm] a12 [mm] δ1 [ ] mn12 [mm]
Values → 2100 2800 200 [-31.33, 31.33] 3
Figure 7.17 shows the modeling of the radius of the pitch curves for driving and driven
non-circular gears r1(δ1) and r2(δ2(δ1)) as a function of the angular displacement δ1.
140
Radiuses of driving and driven gears [mm]
r1(δ1)
r2(δ2(δ1))
120
100
80
60
-40 -20 0 20 40
Input angular displacement δ1 [deg]
Figure 7.17: Radiuses of the pitch curves as a function of input angular displacement δ1
- 132 -
From Figure 7.17 it can be observed that, the radius of non-circular gears produce
variable gear ratio (i12(δ1) ≠ const) and is in depends on the input angular displacement δ1.
When the radius of the driving and driven gears have the same values (for δ1 = 0), then the
gear ratio is i12(δ1) = 1 and the terrain vehicle moves in the straight-ahead direction.
In order to verify the reliability of equation (7.32) for determining the radius of the pitch
curves r1(δ1) and r2(δ2(δ1)), the angular displacement δ2(δ1) of the non-circular gear 2 is
determined from equations (7.30) and (7.31), written by the following expression:
δ1
r1(δ1 )
δ 2 (δ1 ) = ∫ dδ1 . (7.34)
a − r1(δ1 )
0 12
The angular displacement δ2(δ1) of the non-circular gear 2 determined with equation
(7.34) should provide the same values as equation (7.29) in order to exactly follow the
Ackermann steering geometry. Figure 7.18 introduces the angular displacement δ2(δ1) of the
driven non-circular gear 2 as a function of the input angular displacement δ1 of the driving
non-circular gear 1. From Figure 7.18 it can be concluded that, the design of the pair of non-
circular gears follows exactly the Ackermann steering geometry, and will be used for the
steering mechanism of the terrain vehicle.
40 δ2(δ1) - Non-circular gears

Angular displacement δ2(δ1) [deg]
δ2(δ1) - Ackermann
20
0.0
-20
-40
-60
-40 -20 0 20 40
Input angular displacement δ1 [deg]
Figure 7.18: Angular displacements δ2(δ1) obtained from the Ackermann steering geometry
and the non-circular gears’ as a function of angular displacement δ1
Hence, the pitch curves of the driving L1(x1, y1) and driven L2(x2, y2) non-circular gears
in their movable coordinate systems (xi, yi) are determined by the expressions:
x1 = r1 (δ1 ) ⋅ sinδ1 ,
(7.35)
y1 = r1 (δ1 ) ⋅ cosδ1 .
- 133 -
x2 (δ 2 (δ1 )) = r2 (δ 2 (δ1 )) ⋅ sin(δ 2 (δ1 ))

(7.36)
y2 (δ 2 (δ1 )) = r2 (δ 2 (δ1 )) ⋅ cos(δ 2 (δ1 ))
With substitution of the parameters given in Table 7.3, the pitch curves of the driving
and driven non-circular gears in the movable system (xi, yi) is shown in Figure 7.19. From
Figure 7.19 it can be observed that, the distances between the rotation centers O1 and O2 of the
non-circular gears remain constant values and when the movable axes x1 and x2 arise to 0, the
axes y1 and y2 obtain the same values yi = 100 mm, thus the vehicle moves straight-ahead.
200 O2
Movable axis y1 and y2 [mm]

L1 - Driving gear
L2 - Driven gear
150
100
50
O1
0
-100 -50 0 50 100
Movable axis x1 and x2 [mm]
Figure 7.19: Pitch curves of the non-circular gear given by the movable coordinate system
7.3.2.1 Work principles of the steering mechanism with 2WS by using non-circular gears
After modeling the non-circular gears that follow exactly the Ackermann steering geometry,
the design of the steering mechanism for the terrain vehicle with 2WS is developed in this
way. The driver steering efforts transmit from the steering wheel to the steering gearbox
through the steering shaft (propeller shaft).
The design of the steering gearbox consists of a planetary gear, a driving bevel gear,
two driven bevel gears, and two pairs of the non-circular gears. The planetary gear has higher
mechanical efficiency compared with other gear transmissions, and ensures a greater gear
reduction ratio in order to provide drivers with a mechanical advantage for easy turning of the
vehicle. The steering shaft is connected with the sun gear of planetary gear. The carrier of the
planetary gear is connected to the driving bevel gear which splits the driver’s command into
left and right sides, and provides gear reduction. Two driven bevel gears transmit the driver’s
command into left and right driving non-circular gears and offers opposite rotational
direction. This opposite direction of rotation allows for the using two of the same pairs of
non-circular gears for left and right wheels. The driver commands from driving gears are then
transmitted to the driven non-circular gears which precisely follow the Ackermann steering
geometry specified by equation (7.6). The driven non-circular gears of the left and right sides
- 134 -
of the steering gearbox are connected via the propeller shafts to transmit driver command into
bevel gears with ratio 1:1, fitted in the kingpin axle of the left and right steerable wheels. A
typical working principle is shown in Figure 7.20.
Left front Steering propeller Driving and driven Steering propeller Right front
steering wheel shaft non-circular gears shaft steering wheel
Kingpin axle Kingpin axle
Bevel gears, Bevel gears,

Gear ratio: i = 1:1 Gear ratio: i = 1:1
Bevel gears
Gear ratio: i = 2.5:1 Planetary gear
Gear ratio: i = 6:1
Steering shaft
Steering wheel
Non-steerable Non-steerable
left rear wheel right rear wheel
Figure 7.20: Steering mechanism for terrain vehicle with 2WS by using non-circular gears
7.3.3 Design of the vehicle’s steering mechanism with 4WS by using non-circular gears
Designing of the steering mechanism for a terrain vehicle with 4WS is developed in same way
as 2WS in order to exactly ensure the Ackermann steering geometry given by equations (7.15)
and (7.16). The main goal of this steering mechanism is the designing of suitable pairs of the
non-circular gears for front and rear steering wheels. The design principle is similar to that of
vehicles with 2WS but with some substantial differences when determining the pitch curves
of the non-circular gears. In the same ways are determines the angular displacements for the
- 135 -
front δ2F(δ1F) and rear δ2R(δ1R) driven non-circular gears as a function of angular
displacements of the front δ1F and rear δ1R driving non-circular gears, written as:
a cot(cotδ1F − w / c) if − δ1F min ≤ δ1F ≤ 0,

δ 2F (δ1F ) = 
a cot(cotδ1F + w / c) if 0 ≤ δ1F ≤ δ1F max ,
(7.38)
a cot(cotδ1R − w / d ) if − δ1R min ≤ δ1R ≤ 0,
δ 2R (δ1R ) = 
a cot(cotδ1R + w / d ) if 0 ≤ δ1R ≤ δ1R max .
The gear’s ratio for the front i12F(δ1F) and rear i12R(δ1R) for pairs of the non-circular
gears is determined by the following expression:
ω1F (δ1F ) dδ1F / dt dδ1F r2 F (δ 2 F (δ1F ))
i12F (δ1F ) = = = = ,
ω2 F (δ 2 F ) dδ 2 F / dt dδ 2 F r1F (δ1F )
(7.39)
ω (δ ) dδ / dt dδ1R r2 R (δ 2 R (δ1R ))
i12R (δ1R ) = 1R 1R = 1R = = .
ω2 R (δ 2 R ) dδ 2 R / dt dδ 2 R r1R (δ1R )
where ω1F(δ1F), ω2F(δ2F), ω1R(δ1R) and ω2R(δ2R) denotes the angular velocity, whilst r1F(δ1F),
r2F(δ2F(δ1F)), r1R(δ1R) and r2R(δ2R(δ1R)) represent the radius of the pitch curves for front and
rear driving and the driven non-circular gears, respectively.
The distance of the rotation center for the front a12F and rear a12R pairs of the non-
circular gears always remains a constant value, and are determined as follow:
r1F (δ1F ) + r2 F (δ 2 F (δ1F )) = a12F = const,
(7.40)
r1R (δ1R ) + r2 R (δ 2 R (δ1R )) = a12R = const.
The radius of the pitch curves front driving r1F(δ1F) and driven r2F(δ2F(δ1F)) gears and
rear driving r1R(δ1R) and driven r2R(δ2R(δ1R)) gears, are written as follows:
a12F a12R
r1F (δ1F ) = ; r1R (δ1R ) = ,
1 + 1 /(dδ 2 F / dδ1F ) 1 + 1 /(dδ 2 R / dδ1R )
(7.41)
a /(dδ 2 F / dδ1F ) a /(dδ 2 R / dδ1R )
r2 F (δ 2 F (δ1F )) = 12F , r2 R (δ 2 R (δ1R )) = 12R .
1 + 1 /(dδ 2 / dδ1 ) 1 + 1 /(dδ 2 R / dδ1R )
where the derivative functions dδ2F/dδ1F and dδ2R/dδ1R are defined from equation (7.38) and
=
2
[ 2
]
dδ 2 F (δ1F ) 1 / sin δ1F ⋅ 1 + (cotδ1F − w / c) if − δ1F min ≤ δ1F ≤ 0 ,
dδ1F [ ]
1 / sin2 δ1F ⋅ 1 + (cotδ1F + w / c)2 if 0 ≤ δ1F ≤ δ1F max ,
(7.42)
dδ 2 R (δ1R ) 1 / sin δ1R ⋅ [1 + (cotδ − w / d ) ] if − δ
2 2
1R 1R min ≤ δ 1R ≤ 0 ,
=
⋅ [1 + (cotδ + w / d ) ] if 0 ≤ δ
2
dδ1R 1 / sin2 δ1R 1R 1R ≤ δ1R max ,
For the modeling of pairs of the non-circular gears for front and rear steering wheels in
Table 7.4 are given necessary parameters.
- 136 -
Table 7.4: Parameters for modeling non-circular gears for steering system with 4WS
Parameters for modeling non-circular gears

Parameters
w c d a12F a12R δ1F δ1R mn12F mn12R
→ 0 0
[mm] [mm] [mm] [mm] [mm] [] [] [mm] [mm]
Values → 2100 1400 1400 200 200 [-18.5;18.5] [-18.5;18.5] 3 3
The angular displacements for front δ2F(δ1F) and rear δ2R(δ1R) driven non-circular gears
are determined from equations (7.39) and (7.40), and are written as the following expression:
δ1 F δ1 R
r1F (δ1F ) r1R (δ1R )
δ 2 F (δ1F ) = ∫
0
a12F − r1F (δ1F )
dδ1F ; δ 2 R (δ1R ) = ∫a
0 12R − r1R (δ1R )
dδ1R . (7.43)
The angular displacement determined by equation (7.43) should provide the same
values as equation (7.38) in order to follow exactly the Ackermann steering geometry. Figure
7.21.a introduces the angular displacement for the front and rear driven non-circular gears
δ2ForR as a function of the input angular displacement of the front and rear driving non-circular
gears δ1ForR. The front and rear pitch curves of the driving and driven non-circular gears in
their movable coordinate systems (xi, yi) are determined by the expressions:
x1F = r1F (δ1F ) ⋅ sinδ1F ; x1R = r1R (δ1R ) ⋅ sinδ1R ,

(7.44)
y1F = r1F (δ1F ) ⋅ cosδ1F ; y1R = r1R (δ1R ) ⋅ cosδ1R ;
x2F (δ 2F (δ1F )) = r2F (δ2F (δ1F )) ⋅ sin(δ2F (δ1F )); x2R (δ2R (δ1R )) = r2R (δ 2R (δ1R )) ⋅ sin(δ2R (δ1R ))
(7.45)
y2F (δ 2F (δ1F )) = r2F (δ 2F (δ1F )) ⋅ cos(δ2F (δ1F )); y2R (δ2R (δ1R )) = r2R (δ 2R (δ1R )) ⋅ cos(δ2R (δ1R ))
With the substitution parameters shown in Table 7.4, the front and rear pitch curves of
the driving and driven non-circular gears in the (xi, yi) system, are presented in Figure 7.21.b.
40 200
δ2ForR(δ1ForR) – Front or rear non- O2ForR
Angular displacement δ2ForR(δ1 ForR)
Movable axis y1ForR and y2ForR [mm]
circular gears L1 – Front or rear driving gears

δ2ForR(δ1ForR) – Ackermann L2 – Front or rear driven gears
20
condition 150
0
100
-20
50
-40
0 O1ForR
-60
a) -20 -10 0 10 20 b) -100 -50 0 50 100
Input angular displacement δ1ForR [deg] Movable axis x1ForR and x2ForR [mm]
Figure 7.21: The front and rear; a) angular displacements δ2ForR(δ1ForR) of the driven non-
circular gears and b) pitch curves in the (xi, yi) for driving and driven non-circular gears
- 137 -
From Figure 7.21 it can be observed that the characteristics for front and rear pairs of
non-circular gears are the same. These are due to the center of rotation being at the middle of
the terrain vehicle, and the longitudinal distances between the turning center of the front c and
rear d are the same c = d. Furthermore, the pair of non-circular gears exactly follows the
Ackermann steering geometry and is used for the steering mechanism of the terrain vehicle
with 4WS, Figure 7.22.
7.3.3.1 Work principles of the steering mechanism with 4WS by using non-circular gears
After the modeling of the front and rear driving and driven non-circular gears, the steering
mechanism for the terrain vehicle with 4WS conforms exactly to the Ackermann steering
geometry, as illustrated in Figure 7.22. The working principles is the driver through the
steering wheel transmit command to the front steering gearbox via the steering shaft.
Steering propeller Driving and driven Steering propeller

shaft non-circular gears shaft
Left front Right front
steering wheel steering wheel

Bevel gears,
Bevel gears Gear ratio: i = 1:1
Gear ratio: i = 2.5:1
Planetary gear
Steering shaft Gear ratio: i = 6:1
Steering
Steering wheel propeller shaft
Steering
propeller shaft
Bevel gears,
Gear ratio: i = 1:1

Left rear Right rear

Steering Driving and driven Steering
propeller shaft non-circular gears propeller shaft
Figure 7.22: Steering mechanism for the terrain vehicle with 4WS
- 138 -
The design and working principles of the front steering gearbox are similar to the 2WS
system. The substantial differences are in the radiuses of the pitch curves for the pairs of non-
circular gears and on the shaft of the driving non-circular gear a pair of bevel gears is added.
The driver command from the front to the rear steering gearbox are transmitted through two
pairs of bevel gears with gear ratio 1:1, and connected via the propeller shaft.
The rear steering gearbox consists of the driving bevel gear, two driven bevel gears, and
two pairs of non-circular gears. The driving bevel gear with gear ratio 1:1 is activated and the
steering command takes it from the front steering gearbox and split it to the left and right
branches of the rear steering gearbox. The two driven bevel gears transmit the driver’s parts
of into the left and right driving non-circular gears. Then command is followed to the driven
non-circular gears. The driven non-circular gears are connected through the propeller shafts
into the bevel gears with gear ratio 1:1 which are fitted in the kingpin axle of the left and right
rear steerable wheels. This way ensures the turning of the rear wheels in the opposite direction
of the front steerable wheel. This form of steering allows the terrain vehicle to have a higher
rate of maneuverability and successfully avoided the vehicle from steering bumps during
greater vertical wheel motion.
7.3.4 Design of the steering mechanism with 2WS/4WS for terrain vehicle
After designing the steering mechanisms with 2WS and 4WS for terrain vehicles which
exactly follow Ackermann steering geometry, it is necessary to design a steering mechanism
in order that the terrain vehicle can have both systems. By using this steering mechanism a
driver will have the opportunities of deciding whether to drive with the steering of only the
front wheels (2WS - rear wheels are locked) or to steer all the wheels (4WS). The first option
provides greater vehicle turning radius compared to the other option and could be used when
vehicle is driven at higher speed by ensuring better stability of the vehicle. For better
maneuvering in the city or when the vehicle moves over rough terrain, the vehicle’s turning
radius should be as small as possible. This would be ensured by the second option and would
be used when the vehicle moves at lower speed.
7.3.4.1 Work principles of the steering mechanism with 2WS/4WS
The differences between 2WS/4WS and separate 2WS or 4WS are that, both systems are now
fixed in the front steering gearbox. Front steering gearboxes, except gears that also have three
joints fitted (A, B and C). The three joints are connected at the same time by motion towards
the left or right direction by the lever, and don’t have free positions, Figure 7.23. The joints
- 139 -
connect the front driven non-circular gears as 2WS or 4WS and the rear steering gearbox.
Switch of joints could be done mechanically, hydraulically or in an electrical ways. The
switching to 2WS or 4WS modes is selected by the driver and then fixed. Switching to 2WS
or 4WS or reciprocally, depends on the driver’s wish and can only change when the vehicle is
stopped. This is for safety reasons. Figure 7.23 illustrate the unique design of the steering
mechanism with 2WS/4WS modes for terrain vehicles.
Steering propeller Driving and driven Steering propeller

shaft non-circular gears shaft
Left front 4WS 2WS Right front
Bevel gears, A B Bevel gears,

C
Bevel gears,
Bevel gears Gear ratio: i = 1:1
Gear ratio: i = 2.5:1
Planetary gear
Steering shaft Gear ratio: i = 6:1
Steering
Steering wheel propeller shaft
Steering
propeller shaft
Bevel gears,
Gear ratio: i = 1:1
Bevel gears,
Bevel gears,
Gear ratio: i = 1:1
Gear ratio: i = 1:1
Kingpin axle
Kingpin axle
Right rear
Left rear
steering wheel
steering wheel Steering Driving and driven Steering
propeller shaft non-circular gears propeller shaft
Figure 7.23: Steering mechanism of the terrain vehicle with 2WS/4WS
If the joints (A, B and C) moves in the left direction steering system provide 4WS
mode, otherwise ensure 2WS mode.
- 140 -
7.4 Discussion of results
Designing a steering mechanism for terrain vehicles was done for front wheel steering and all
wheel steering. Designs are in relation to all the requirements given in the check list. In all
operation modes of the steering mechanism the target was a system to provide exactly the
Ackermann steering geometry.
The trapezoid steering mechanism conforms closely to the Ackermann conditions by
providing a steering error from e = 1.30. On the one hand, the characteristic of the terrain
vehicle is largely vertical wheel movements due to protecting the vehicle from rollover risk.
On the other hand, these movements cause so-called bump steering by cause a reduction in
the vehicle stability. A better solution was thus sought. By employing a pair of non-circular
gear modeled according to the Ackermann principle ensures the Ackermann steering
geometry and avoids steering bumps. In this situation the steering wheels during all vertical
motion will stay in the same position without minor rotation around its kingpin axis.
The design of this steering system with non-circular gears was done for three cases:
2WS, 4WS, and 2WS/4WS. All cases ensure Ackermann steering conditions. Differences in
performance that provide 4WS compared with 2WS is the vehicle turning radius reduction of
34%. This means that by this system the terrain vehicle can demonstrate higher
maneuverability when moving in city or over rough terrain. By utilizing 2WS/4WS vehicle
performance significantly improves at both high and low speeds.
- 141 -
8 SUMMARY AND CONCLUSION
After the known research problems regarding the suspension and steering system were
identified, three key elements became the main objectives of this research project.
The first objective the developing of an optimal suspension mechanism for a terrain
vehicle with four wheels-drive and four-wheels steering. The second goal was to design and
optimize control strategies for a suspension system that would minimize undesired vibration
that affects a vehicle’s body caused by ground excitation, and provide better driving comfort.
The third topic was to develop an effective steering system that would enable the provision of
good maneuverability at lower speeds and efficient driving safety at higher speeds.
Obtaining these objectives was carried out by more or less comprehensive systems. The
known suspension and steering systems are developed sufficiently enough for passenger cars
but when a terrain vehicle is in question with greater wheel motions due to rollover risk, the
situation is totally different.
The first goal is reached by designing a completely new suspension mechanism which
was derived by improving a double wishbone with longer equal control arms. Mathematical
modeling and solving of equations of motion had been developed by numerical methods for a
quarter of it. In order to provide less lateral wheel motion, the Finite Element Methods were
employed in order to optimize. The following summarizes the main features concerning the
developed suspension mechanism:
• When the wheel moves from its rest position to the rebound direction, the
suspension mechanism provides 45 % less lateral wheel motion, whilst when the
wheel moves from the rest to the bound direction it obtains better behavior by
ensuring 72.6 % less when compared with a predesigned suspension mechanism
with double wishbone and a longer equal control arm,
• The suspension mechanism provides zero camber angles during all vertical wheel
motion such as the rebound and bound directions, and
• Its optimized suspension mechanism provides up to 37.5% less lateral wheel motion
when compared with the initially developed suspension mechanism.
The proposed suspension design provides relatively small lateral wheel motion, zero
camber angles, and effectively absorbs the vibrations caused by ground excitation. These
- 142 -
features affect the improvement of tire life-expectancy, increasing the quality of the tire
contact with the road surface, thus reducing the additional load in the suspension mechanism.
The second objective of this approach is obtained by design and optimization control
strategies for active, semi-active, and passive suspension systems. A simplified quarter
vehicle model was introduced, together with a suitably-created mathematical model for active,
semi-active and passive systems. Solving of the differential equations of motion was carried
out within a MATLAB/Simulink environment. The ground excitation was modeled using two
repeated cosine bumps and flat lines. Optimal suspension parameters for a passive linear
suspension system were optimized by using Multi Objective Genetics Algorithms. Whilst for
optimal parameters for the active damping forces were determined by the stochastic
parametric optimization method, which is based on non-linear programming, the suspension
performances were optimized by maximizing the driving comfort and safety. The obtained
results were gained by solving differential equations of motion for active, semi-active, and
passive non-linear systems. The following features were identified concerning the developed
control strategy:
• Vehicle’s body acceleration for optimal design of the passive linear suspension
system in some locations is reduced by 75%, whilst the vertical tire forces by 71%
less compared with the initial design of the passive linear suspension system,
• Vehicle’s body acceleration for optimal design of the actives non-linear suspension
system in some locations is reduced by 81%, whilst the vertical tire forces by 54%
less compared with the passive non-linear suspension system, and
• Vehicle’s body acceleration for optimal design of the semi-active non-linear
suspension system in some locations is reduced by 76%, whilst the vertical tire
forces by 49% less compared with the passive non-linear suspension system.
It is obvious that the active system provides better performances compared with the
semi-active and passive ones.
The third aim has been to developing an effective steering system for terrain vehicle
with four wheel steer. This steering mechanism is designed to work by two modes steer such
as all wheels steer and front wheels steer. Proposed steering mechanism conform exactly the
Ackermann steering geometry in all modes and situations.
The trapezoid steering mechanism conforms closely with the Ackermann law by
providing 1.30 steering error in some steering angles. This error will probably not be a
problem if the terrain vehicle moves over flat roads but problems will appear when it moves
over bumps or rough terrain such as bump steering. Effective steering design consists of
- 143 -
special pairs of gears known as non-circular gears. The following features can be identified
regarding the new steering mechanism:
• The vehicle turning radius for all wheel steering is reduced by 34% compared with
only front wheel steering,
• The Ackermann steering condition is ensured 100%, and
• Bumps steering caused when the vehicle moves over rough terrain can be perfectly
avoided.
So, the proposed steering system for the terrain vehicle demonstrates good
maneuverability at low speed and provides suitable stability at higher speeds due to following
Ackermann law.
Finally, on the basis of comprehensive analysis shown during this doctoral thesis, it can
be concluded that the developed suspension and steering system for terrain vehicles with four
wheel drives and four wheel steering can ensure efficient driving comfort, driving safety, and
good maneuverability. In addition, this research project contributes to the state of knowledge
for the designing of suspension and steering systems whilst building bridges between
designers, manufactures, researchers, and so one. Furthermore, this approach has already
resulted in several academic publications given follows:
[1] Lajqi Shpetim, Pehan Stanislav, Lajqi Naser, Gjelaj Afrim, Pšeničnik Jože, Sašo Emin.
Design of Independent Suspension Mechanism for a Terrain Vehicle with Four Wheels Drive
and Four Wheels Steering. Annals of Faculty Engineering Hunedoara - International Journal
of Engineering/Fascicule1 (2013), Fascicule1, Tome XI 2013, pp. 101-108.
[2] Lajqi Shpetim, Stanislav Pehan. Designs and Optimizations of Active and Semi-Active Non-
linear Suspension System for a Terrain Vehicle. Strojniški vestnik - Journal of Mechanical
Engineering (2012), vol. 58, no.12, pp. 732-743.
[3] Lajqi Shpetim, Gugler Jürgen, Lajqi Naser, Shala Ahmet, Likaj Ramë: Possibilities
Experimental Method to Determine the Suspension Parameters in a Simplified Model of a
Passenger Car. Inter. Journal of Automotive Technology (2012), vol. 13, no. 4, pp. 615-621.
[4] Lajqi Shpetim, Pehan Stanislav, Lajqi Naser, Baxhaku Bashkim, Ymeri Besim. Influences of
the Suspension Parameters on the Vehicle Suspension Performance for a Terrain Vehicle.
Journal of Mechanics Engineering and Automation (2012), vol. 2, no. 9, pp. 550-554.
[5] Pehan Stanislav, Lajqi Shpetim, Pšeničnik Jože, Flašker Jože. Modeling and Simulation of off
Road Vehicle with Four Wheel Steering. The 7th International Scientific Conference,
Research and Development of Mechanical Elements and Systems (2011), IRMES 2012
Conference Proceedings, pp. 77-83.
[6] Lajqi Shpetim, Pehan Stanislav, Pšeničnik Jože, Lajqi Naser. Tracking the Prescribed Path
Through an Autonomous Mobile Robot with Four Wheel Steering. The 3th International
Scientific Conference, Management of Technology Step to Sustainable Production (2011),
MOTSP 2011 Conference proceedings, pp. 94-101.
- 144 -
LIST OF USED SOURCES
[1] Pehan Stanislav, Lajqi Shpetim, Pšeničnik Jože, Flašker Jože. Modeling and simulation
of off road vehicle with four wheel steering. IRMES 2012 Conference Proceedings
(2011), pp. 77-83.
[2] Lajqi Shpetim, Stanislav Pehan. Designs and Optimizations of Active and Semi-
Active Non-linear Suspension System for a Terrain Vehicle. Strojniški vestnik -
Journal of Mechanical Engineering (2012), vol. 58, no.12, pp. 732-743.
[3] Lajqi Shpetim, Pehan Stanislav, Lajqi Naser, Gjelaj Afrim, Pšeničnik Jože, Sašo
Emin. Design of Independent Suspension Mechanism for a Terrain Vehicle with Four
Wheels Drive and Four Wheels Steering. MOTSP 2012 Conference Proceedings
(2012), pp. 230-237.
[4] Rill Georg. Vehicle dynamics. Regensburg University of Applied Science. Regensburg,
2009.
[5] http://www.interregs.com/spotlight.php?id=27/[17.08.2011].
[6] RTC - Zavod za avtomobilsko industrijo (Razvojno Tehnološki Center)
(http://www.rtc.si), Ceska k TAMU 7, Maribor, 2011&2012.
[7] Genta Giancarlo, Morello Lorenzo. The Automotive Chassis-Volume 1: Components
Design. Springer, Torino, 2009.
[8] Dixon C. John. Suspension Geometry and Computation. John Wiley, New York, 2009.
[9] Jazar Reza. Vehicle dynamics: Theory and Application. Springer, New York, 2009.
[10] http://www.daimler.com/dccom/0-5-1324882-49-1324893-1-0-0-1345593-0-0-135-0-0-
0-0-0-0-0-0.html/[01.02.2012].
[11] Mohd Zakaria Mohammad Nasir. Analysis of double wishbone suspension system
(2008). Master dissertation in Automotive Engineering. Coventry University, School
Engineering.
[12] Gillespie Thomas. Fundamental of Vehicle Dynamics. Society of Automotive
Engineering. Warrendale, 1992.
[13] Duffy James, Krick Russell. Modern Automotive Technology-Suspension System
Fundamental. Goodheart-Wilcox Publisher Co., Illinois, 2004, pp. 1280-1301.
[14] Eslaminasab Nima. Development of a Semi-active Intelligent Suspension System for
Heavy Vehicles (2008). Doctoral thesis, University of Waterloo, Waterloo.
- 145 -
[15] Fisher, R., Gscheidle, R., Heider, U., Hohmann , B., Keil, W., Mann, J., Schlogl, B.,
Wimmer, A., Wormer, G. Fachkunde Kraftfahrzeugtechnik, Europa Lehrmittel,
Germany, 2009.
[16] Güler Duygu. Dynamic analyses of double wishbone suspension (2006). Master thesis.
Izmir Institute of Technology, Izmir.
[17] http://www.autoevolution.com/news/how-multi-link-suspension-works-7804.html/
[28.01.2012].
[18] Lajqi Shpetim, Gugler Jürgen, Lajqi Naser, Shala Ahmet, Likaj Ramë: Experimental
possibilities for determination of suspension parameters by the simplified model of
passenger car. International Journal of Automotive Technology (2012), vol. 13, no. 4,
pp. 615-621.
[19] Senthil kumar Mouleeswaran. Genetic Algorithm – Based Proportional Derivative
Controller for the Development of Active Suspension System. Information Technology
and Control (2007), vol. 36, No. 1, pp. 58-67.
[20] Wong Jo Yung. Theory of Ground Vehicle. John Wiley & Sons, Inc., New York, 2001.
[21] Nekoui A. Mohammad, Hadavi Parisa. Optimal Control of an Active Suspension
System. 14th International Power Electronics and Motion Control Conference (2010),
T5-60-T5-63.
[22] Karnopp D. C., Crosby M. J., Harwood R. A. Vibration Control Using Semi-Active
Force Generators. ASME Journal of Engineering for Industry (1974), vol. 96, no. 2,
pp. 619-626.
[23] Pajaziti Agron. Contribution in the Parametric Optimization of the Shock Absorber
Characteristic with Regulation in Aspect of the Vibration Impact in the Comfort and
Stability to the Passenger Vehicles (1992). Doctoral thesis. University of Prishtina,
Prishtina.
[24] Lin Jung-Shan, Kanellakopoulos Ioannis. Nonlinear design of active suspensions. IEEE
Control Systems Magazines (1997), vol. 17, no. 3, pp.45-59.
[25] Yi Kyongsu, Song S. Byung. Observer Design for Semi-active Suspension Control,
Vehicle System Dynamics (1999), vol. 32, pp. 129-148.
[26] Hong, K. S., Sohn H. C., Hedrick J. K. Modified Skyhook Control of Semi-active
Suspension: A New Model, Gain Scheduling, and Hardware-in-the Loop Tuning. ASME
Journal of Dynamics System, Measuring and Control (2002), vol. 124, no. 1,
pp. 158-167.
[27] Turnip Arjon, Hong Keum-Shik, Park Seonghun. Control of a Semi-Active MR-
- 146 -
Damper Suspension System: A New Polynomial Model. Proceedings of the 17th Word
Congress. The International Federation of Automatic Control (2008). Seoul, Korea,
pp. 4683-4688.
[28] Erjavec Jack. Automotive Technology-A systems Approach: 5th Edition. Delmar,
Cengage Learning, 2010.
[29] Mokhlespour Esfahani Mohammad Iman, Mosayebi Masoud, Pourshams Mohammad,
Keshavarzi Ahmad. Optimization of Double Wishbone Suspension System with
Variable Camber Angle by Hydraulic Mechanism, World Academy of Science,
Engineering and Technology (2010), vol. 61, pp. 299-306.
[30] Allred M. Timothy. Compliant mechanism system (2003). Master thesis. Brigham
Young University.
[31] Heo J-H, Lee U-K, Lee S-H. Development of a method to compute the kingpin axis
using screw axis theory based on suspension-parameter-measuring device data.
Proceeding of the Institution of Mechanical Engineers, Part D: Journal Automobile
Engineering (2009), vol. 223, no. 4, pp.: 519-531.
[32] Maher Damien. Combined Time and Frequency Domain Approaches to the Operational
Identification of Vehicle Suspension Systems (2011). Doctoral thesis. Dublin City
University, School of Mechanical and Manufacturing Engineering, Dublin 9.
[33] Bouazara Mohamed, Richard Marc. An optimization method designed to improve 3-D
vehicle comfort and road holding capability through the use of active and semi-active
suspension. European Journal of Mechanic-A/Solids (2001), vol. 20, no. 3, pp. 509-520.
[34] Gobbi M., Mastinu G. Analytical description and optimization of the dynamic behavior
of passively suspended road vehicles. Journal of Sound and Vibration (2001), vol. 245,
no. 3, pp. 457-481.
[35] Baumal A.E., McPhee J.J., Calamai P.H. Application of Genetic Algorithms to the
Design Optimization of an Active Vehicle Suspension System. Computer Methods in
Applied Mechanics and Engineering (1998), vol. 163, pp. 87-94.
[36] Alkhatib R., Jazar G.N., Golnaraghi M.F. Optimal Design of Passive Linear Suspension
Using Genetic Algorithm. Journal of Sound and Vibration (2004), vol. 275,
pp. 665-691.
[37] Yu Hong Biao, Yu Nan. Application of Genetic Algorithms to Vehicle Suspension
Design (2003). The Pennsylvania State University, University Park, pp. 1-9.
[38] Heisler Heinz. Advanced Vehicle Technology. Butterworth Heinemann, Woburn, 2002,
pp. 654.
- 147 -
[39] http://www.scribd.com/doc/34677412/mechanical-four-wheels-steering/[22.11.2011].
[40] Alter M. David. Transient and Steady-State Performance Comparison of Two-Wheel-
Steer and Four-Wheel- Steer Vehicle Models. Department of Mechanical and Industrial
Engineering, University of Illinois at Urbana Champaign, Illinois, 1989, pp. 787-792.
[41] Spentzas K., Alkhazali I., Demic M. Kinematic of four-wheel-steering vehicles.
Forschung in Ingenieurwesen 66 (2001), Springer-Verlang, pp. 211-216.
[42] Brabec P., Maly M., Vozenilek R. Controls system of vehicle model with four wheel
steering (4WS), International Scientific Meeting Motor Vehicles & Engines, Kragujevac
(2004), pp. 7.
[43] Lee Jong-Nyun. Design sensitivity analysis of multi-body system with special
references to four wheel steering. Department of Aerospace and Mechanical
Engineering, University of Arizona, 2002, pp. 159.
[44] Emura Takashi, Arakawa Akira. A New Steering Mechanism Using Noncircular Gears,
JSME International Journal, (1992), vol. 35, no. 3, pp. 604-610.
[45] Pahl Gerhrd, Beitz Wolfgang, Feldhusen Jörg, Grote Kark-Heinrich. Engineering
Design: A systematic Approach. Springer – 3rd edition, Berlin, 2007.
[46] Mastinu Giampiero, Gobbi Massimiliano, Miano Carlo. Optimal design of complex
Mechanical Systems: With Application to Vehicle Engineering. Springer, London,
Berlin, 2006.
[47] Balič Jože. Contribution to integrated manufacturing. DAAAM Publishing series,
Vienna, 1999.
[48] Mecitoglu Zahit. Finite Element Analysis in Structures. Faculty of Aeronautics and
Astronautics, Maslak, Istanbul, 2008.
[49] Sorić Jurica. Metoda Konačnih Elemenata. Sveučilište u Zagrebu, Zagreb, 2004.
[50] Mackerle Jaroslav. FEM and BEM in the context of information retrieval. Computer
and Structures (2002), vol. 80, no. 20-21, pp. 1595-1604.
[51] Roylance David. Finite Element Analysis. Massachusetts Institute of Technology,
Cambridge, 2001, pp. 1-16.
[52] Lenart Sašo. Optimiranje oblike nastavnih čeljusti ročnih klešč. Magistersko delo.
Universa v Maribory, Fakulteta za stroništvo, Maribor, 2009.
[53] Likaj Ramë, Shala Ahmet, Bruqi Mirlind, Qelaj Mehmet. Optimal design of quarter car
vehicle suspension system. TMT 2010 Conference Proceedings (2010), pp. 417-420.
[54] Reimpell Jörsen, Stoll Helmut, Betzler Jürgen. The automotive Chassis: Engineering
Principles. Butterworth-Heinemann, London, 2001.
- 148 -
[55] Werff Klaas. Kinematic and dynamic analysis of mechanisms, a finite element
approach. Technical University of Delft, Nederland, 1977.
[56] http://www.workingmodel.com / Working Model Software.
[57] Happian-Smith Julian. An introduction to Modern Vehicle Design. Butterworth –
Heinemann, Woburn, 2002.
[58] http://www.ptc.com/go/mathsoft / MathCAD Software
[59] Blundell Mike, Harty Damian. The Multibody Systems Approach to Vehicle Dynamics.
Elsevier Butterworth-Heinemann, UK, 2004.
[60] Popa Elena Cristina. Steering System and Suspension Design for 2005 Formula SAE-A
Racer Car. University of Southern Queensland - Faculty of Engineering and Surveying,
Australia, 2005.
[61] Kegel M., Butinar B., Kegel B. An efficient gradient – based optimization algorithm for
mechanical systems. Commun. Numer. Meth. Eng. (2002), vol. 18, pp. 363-371.
[62] Pehan Stanislav, Kegl Breda, Pogorevc Primož. Developing a Racing Car. Journal of
Mechanical Engineering (2003), vol. 49-12, pp. 593-606.
[63] Kegl Breda, Marko Kegl, Stanislav Pehan. Optimization of a Fuel Injection System for
Diesel and Biodiesel Usage. Energy & Fuel (2008), 22, pp. 1046-1054.
[64] Kegl M., Antes H. Shape optimal design of elastic space frames with non linear
response. Int. J. Numer. Meth. Eng. (1998), vol. 43, pp. 93-110.
[65] Kegl M., Brank B. Shape optimization of truss-stiffened shell structures with variable
thickness. Comput. Methods Appl. Mech. Eng. (2006), vol. 195, pp. 2611-2634.
[66] Zidarič Ivan. Numerično optimiranje nosilnega okvira dirkalnega vozila. Magistersko
delo. Universa v Mariboru, Fakulteta za stroništvo, Maribor, 2003.
[67] STAKx-Version 2009.09.16. Copyright © by: Marko Kegl, Boštjan Harl. Faculty of
Mechanical Engineering, University of Maribor & Boštjan Brank, Matija Gams. Faculty
of Civil and Geodetic Engineering, University of Ljubljana.
[68] iGOx – Version 2009.09.16. Copyright © by: Marko Kegl, Boštjan Harl. Faculty of
Mechanical Engineering, University of Maribor.
[69] Belingardi Giovanni, Demic Miroslav. A Contribution to Shock Absorber Modeling
by Using “Black Box” Method. Scientific Bulletin of Faculty of Mechanical and
Technology, Automotive series (2009), vol. 19, no. B, pp. 1-15.
[70] Demic M., Diligenski D., Demic I., Demic M. A method of active suspension design.
Forsch Ingenieurwes (2006), vol. 70, pp. 145-158.
- 149 -
[71] Ram Mohan Rao T., Venkata Rao G., Sreenivasa Rao K., Purushottam A. Analysis of
Passive and Semi-active Controlled Suspension Systems for Ride Comfort in an
Omnibus Passing Over a Speed Bump. International Journal of Research and Reviews
in Applied Science (2010), vol. 5, no. 1, pp. 7-17.
[72] Demic M. Optimization of the Oscillation Systems in Motor Vehicles. Faculty of
Mechanical Engineering, Kragujevac, 1997.
[73] Abramov Sergey, Mannan Samjid, Durieux Olivier. Semi-Active Suspension System
Simulation Using Simulink. International Journal of Engineering System. Modeling
and Simulation (2009), vol. 1, no. 2/3, pp. 101-114.
[74] Sam Y. M., Osman J. H. S., Ghani M. R.A. Active Suspension Control: Performance
Comparison using Proportional Integral Sliding Mode and Linear Quadratic Regulator
Methods. IEEE (2000), pp. 274-278.
[75] Lajqi Shpetim, Pehan Stanislav, Lajqi Naser, Baxhaku Bashkim, Ymeri Besim.
Influences of the Suspension Parameters on the Vehicle Suspension Performance for a
Terrain Vehicle. Journal of Mechanics Engineering and Automation (2012), vol. 2,
no. 9, pp. 550-554.
[76] Popp Karl, Schiehlen Werner, Ground Vehicle Dynamics. Springer, Verlang, Berlin
Heidelberg, 2010.
[77] Zhang D. Kinetostatic analysis and optimization of parallel and hybrid architectures for
machine tools (2000), Doctoral thesis, University of Laval Quebec.
[78] Fundamental of Genetic Algorithms: AI Course Lecture 39 – 40, (2012) pp. 42,
Available at http://www.myreaders.info/html/artificial_intelligence.html/25/04/12.
[79] Schaffer JD. Multiple objective optimization with vector evaluated genetic algorithms.
Proceedings of the International Conference on Genetic Algorithm and their
applications (1985), pp. 93-100.
[80] Fonseca CM., Fleming PJ. Multi-objective genetic algorithms. The Institution of
Electric Engineers, IEE (1993), London, pp. 1-5.
[81] Kuznestov Alexey, Mammadov Musa, Sultan Ibrahim, Hajilarov Eldar. Optimization
of improved suspension system with inerter device of the quarter-car model in
vibration analyses. Archive of Applied Mechanics (2011), vol. 81, no. 10,
pp. 1427-1437.
[82] http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=CELEX:31999L0007:en:NOT
/[12.09.2012].
- 150 -
[83] Leishman C. Robert. Application of Variations Analysis Methods to Automotive

Mechanisms (2009), Master thesis, Brigham Young University.
[84] Alanoly J., Sankar S. A new concept in semi-active vibration isolation. American
Society of Mechanical Engineers, Journal of Mechanics, Transmissions and Automation
in Design (1987), pp. 242-247.
[85] Adam Theander. Design of a Suspension for Formula Student Race Car (2004). Master
thesis. Vehicle Dynamic Aeronautic and Vehicle Engineering Royal Institute of
Technology, Stockholm.
[86] Lajqi Shpetim, Pehan Stanislav, Pšeničnik Jože, Lajqi Naser. Tracking the prescribed
path through an autonomous mobile robot with four wheel steering. MOTSP 2011
Conference Proceedings (2011), pp. 94-101.
[87] Khurmi R., Gupta J. Theory of Machines. S Chand & Co Ltd, India, 2005.
[88] Dormand, J. R., Prince P. J. A family of embedded Runge-Kutta formulae. Journal
Computational Applied Mathematics (1980), vol. 6, no. 1, pp. 19-26.
[89] Lu Ya Yan. Numerical Methods for Differential Equations. Department of
Mathematic, City University of Hong Kong, Kowloon, Hong Kong, 2011.
[90] Konak Abdullah, Coit W. David, Smith E. Alice. Multi-objective optimization using
genetic algorithms: A tutorial. Reliability Engineering & System Safety (2006), 91,
pp. 992-1007.
[91] Zhao Sheng, Zhao Jing-Shan, Feng Zhi-Jing. Design of a Pair of Noncircular Gears
Meeting Ackermannn Steering Principle. IEEE (2011), pp. 604-610.
[92] Guglielmino E, Sireteanu T., Stammers Ch., Ghita G., Giuclea M. Semi – active
Suspension Control-Improved Vehicle Ride and Road Friendliness: Published by
Springer. London, UK, 2008, pp. 294.
[93] Ginsberg H. Jerry. Advanced engineering dynamics. Published by Harper & Row. Ney
York, USA, 1988, pp. 432.
[94] Grujicic M., Arakere G., Bell W.C., Marvi H., Yalavarthy H.V., Pandurangan B.,
Haque I., Fadel G.M. Reliability - Based Design Optimization for Durability of Ground
Vehicle Suspension System Components. Journal of Materials Engineering and
Performance (2010), vol. 19, pp. 301-313.
[95] Hao W., Shaoping W., Mileta M. T. Suspension mechanism simulation and
optimization considering multiple disciplines. Journal of Mechanical Science and
Technology (2009), vol. 23, pp. 100-108.
- 151 -
[96] Grujicic M., Bell C W., Arakere G. Haque. Finite element analysis of the effect of up-
armoring on the off-road braking and sharp-turn performance of a high-mobility multi-
purpose wheeled vehicle. Journal Automobile Engineering (2009), vol. 223, Part D,
pp. 1419-1434.
[97] Pfeffer P. E., Harrer M., Johnston D. N. Interaction of vehicle and steering system
regarding on-centre handling. Vehicle System Dynamics (2008), vol. 46, no. 5,
pp. 413-428.
[98] Uys P.E., Els P.S., Thoresson M. Suspension settings for optimal ride comfort of off-
road vehicles travelling on roads with different roughness and speeds, Department of
Mechanical and Aeronautical Engineering, University of Pretoria, Pretoria, 2007.
[99] Pehan S., Kegl B. Top race car development. V: DAŠIĆ, Predrag (ur.). RaDMI 2007:
proceedings on CD-ROM. Kruševac: Institute IMK "14. October"; Trstenik: High
Technical Mechanical School, 2007, str. 27-35.
[100] Hiroyasu O., Masashi T., Nobuaki K., Akira I., Katsuya I. Control for Four – Wheel
Individual Steering and Four – Wheel Driven Electronic Vehicle. Electrical
Engineering in Japan (2005), vol. 153, no. 3, pp. 71-78.
[101] Pehan S., Kegl M., Vipavc D. Optimizacija položaja zgibov zadnje obese kolesa. V:
Fajdiga, Matija (ur.), Klemenc, Jernej (ur.), Kostanjevec, Andrej (ur.). 7. konferenca in
razstava Inovativna avtomobilska tehnologija IAT'05 Bled (2005), Zbornik referatov,
(IAT, Inovativna avtomobilska tehnologija). Ljubljana: ULJ-FS-LAVEK: ZSIS-SVM,
2005, str. 367-378.
[102] Kegl B., Pehan S. Computer aided development of a racing car. V: Budin, Leo (ur.),
Ribarić, Slobodan (ur.). Computers in technical systems. Intelligent systems:
proceedings of the joint conferences CTS+CIS: zbornik radova zajedničkih savjetovanja
CTS+CIS. Rijeka: Hrvatska udruga za mikroprocesorske, procesne i informacijske
sustave, mikroelektroniku i elektroniku - MIPRO HU, 2004, str. 75-81.
[103] Bian X. L., Song B. A., Walter R. Optimization of steering linkage and double-
wishbone suspension via R-W multi-body dynamic analysis, Forschung im
Ingenieurwesen (2004), vol. 69, Springer-Verlag, pp. 38-43.
[104] Pehan S., Kegl B. The modern race car suspension. V: MVM 04. Kragujevac: [s.n.],
2004, str. 381-392, MVM04-A52.
- 152 -
[105] João P. G., Jorge C. A.: Optimization of Vehicle Suspension Systems for Improved
Comfort of Road Vehicles Using Flexible Multi body Dynamics, Nonlinear Dynamics
34: 2003. Kluwer Academic Publishers, Printed in the Netherlands, pp. 113-131.
[106] Zidarič I., Pehan S., Kegl M.: Numerično optimiranje šasije dirkalnega vozila FS-03. V:
Fajdiga, Matija (ur.), Klemenc, Jernej (ur.), Kostanjevec, Andrej (ur.), TRENC,
Ferdinand (ur.). 6. konferenca IAT'03, Koper/Portorož, Slovenija, 24.-25. april 2003.
Zbornik referatov, (IAT, Inovativna avtomobilska tehnologija). Ljubljana: ZSITS,
SVM: Fakulteta za strojništvo, LAVEK, 2003, str. 387-394.
[107] Pehan, Stanislav, Kegl, Breda, Pogorevc, Primož. Oblikovanje dirkalnika - Developing
a racing car. Stroj. vestn. (2003), letn. 49, št. 12, str. 593-606.
[108] Pehan S., Kegl B.: Design of formula student racer. V: Marjanović, Dorian (ur.).
Proceedings of the 6th International Design Conference DESIGN (2000), Cavtat,
Dubrovnik, Croatia, Zagreb: Centre of Technology Transfer: Faculty of Mechanical
Engineering and Naval Architecture; Zürich: Workshop Design-Konstruktion, 2000, str.
483-488.
[109] Madhusudan R.: Number and dimensional synthesis of independent suspension
mechanisms. Mech. Mach. Theory (1996), vol. 31, no. 8, pp. 1141-1153.
[110] Pehan S.: Metodika konstruiranja. Fakulteta za strojništvo, 2005. X, 218 str., graf.
prikazi. ISBN 86-435-0710-5.
[111] Tuan – Ahn N.: Application of Optimization Methods to Controller Design for Active
Suspensions. Doctoral Thesis. Cottbus, Germany, 2006.
[112] Michael J. Th.: Mathematical Optimization of the Suspension System of an Off Road
Vehicle for Ride Comfort and Handling, University of Pretoria, November 2003,
Master Thesis, pages 110.
- 153 -
Appendixes
- 154 -
Appendix A: Free Body Diagrams and Forces That Act in Mechanism
The free body diagrams and forces that act in components of the suspension mechanism are
presented in Figure A.1, Figure A.2 and Figure A.3.
a) FzA b) FzC2 θ (t)

θ1(t) θ9(t) 2
Fzsd
z1
A FyA z2 Fsd FzE C FyC2
az1
Fysd
ε1(t) az2
1 C1 2 C2 E FyE
ay1 y1
FzC4 ε2(t) ay2 y2
m1⋅g
m2⋅g
FyC2 C
FzC2 FyH H
Body 1 FzH Body 2
Figure A.1: Free - body diagram; a) Body 1, and b) Body 2
Dynamic equation of motion for Body 1 – Revolute joints A & C (Figure A.1.a):
m1 ⋅ ay1 = ∑Fyi ; ⇒ m1 ⋅ ay1 = FyA − FyC2 , (A.1)
m1 ⋅ az1 = ∑Fzi ; ⇒ m1 ⋅ az1 = FzA − FzC2 − FzC4 − m1 ⋅ g , (A.2)

+ CG
J c1 ⋅ ε1 = ∑M x1 ; ⇒ J c1 ⋅ ε1 (t ) = FyA ⋅ b1 − FzA ⋅ b2 + FyC 2 ⋅ b1 − FzC 2 ⋅ b2 − FzC 4 ⋅ b2 , (A.3)
when a1 and a2 are substitution by expressions:

AC AC
b1 = b1 (t ) =⋅ sin(θ1 (t ) ) ; b2 = b2 (t ) = ⋅ cos(θ1 (t ) ) . (A.4)
2 2
Dynamic equation of motion for Body 2 – Revolute joints C, E & H (Figure A.1.b):
m2 ⋅ ay2 = ∑Fyi ; ⇒ m2 ⋅ ay2 = FyC2 + FyE + Fysd − FyH , (A.5)
m2 ⋅ az2 = ∑Fzi ; ⇒ m2 ⋅ az2 = FzC2 + FzE − Fzsd − FzH − m2 ⋅ g , (A.6)

+ CG
J c 2 ⋅ ε 2 = ∑ M x 2 ; ⇒ J c 2 ⋅ ε 2 (t ) = FyH ⋅ c1 − FzH ⋅ c2 + FyC 2 ⋅ c1 L
(A.7)
− FzC 2 ⋅ c2 + FyE ⋅ c3 + Fysd ⋅ c3 + Fzsd ⋅ c4 − FzE ⋅ c4 .
when b1, b2, b3 and b4 are substitution by expressions:
CH CH
c1 = c1 (t ) = ⋅ sin(θ 2 (t ) ) ; c2 = c2 (t ) = ⋅ cos(θ 2 (t ) ) , (A.8)
2 2
- 155 -
 CH   CH 
c3 = c3 (t ) =  − CE  ⋅ sin(θ 2 (t ) ) ; c4 = c4 (t ) =  − CE  ⋅ cos(θ 2 (t ) ) . (A.9)
 2   2 
Referring back to the discussion of spring & damper forces given by expressions in
chapter 5, again remind that for a linear formulation based on the spring stiffness (ks), length
of spring (Ls) and the shock absorber damping coefficient (csh), can formulate the spring &
damper forces (Fsd), which is given by expressions:
Fysd = Fsd ⋅ cos(θ 9 (t ) ) = {k s ⋅ [Ls − Ls ( def )] − csh ⋅ v Ls }⋅ cos(θ 9 (t ) ) ,

(A.10)
Fzsd = Fsd ⋅ sin(θ 9 (t ) ) = {k s ⋅ [Ls − Ls ( def )] − csh ⋅ v Ls }⋅ sin(θ 9 (t ) ) .
a) b) FzC4 c) FzB
θ4(t) θ5(t)
FzG
z3 z5
G FyG az3 C az5 B FyB
z4
θ3(t) ay3 y3 4 az4 ε4(t) 5 ε5(t)
C3 a y a y
3 ε3(t) C4 y4 4 C5 y5 5
E m4⋅g FzD4
m3⋅g FyE m5⋅g
D
FzE FyD6 D
FzD4
FzD6
Body 3 Body 4 Body 5
Figure A.2: Free - body diagram; a) Body 3, b) Body 4, and c) Body 5
Dynamic equation of motion for Body 3 – Revolute joints E & G (Figure A.2.a):
m3 ⋅ ay3 = ∑Fyi ; ⇒ m3 ⋅ ay3 = FyG − FyE , (A.11)
m3 ⋅ az3 = ∑Fzi ; ⇒ m3 ⋅ az3 = FzG − FzE − m3 ⋅ g , (A.12)

+ CG
J c3 ⋅ ε 3 = ∑M x3 ; ⇒ J c3 ⋅ ε 3 (t ) = FyG ⋅ d1 + FzG ⋅ d 2 + FyE ⋅ d1 + FzE ⋅ d 2 . (A.13)
when c1 and c2 are substitution by expressions:

EG EG
⋅ sin(θ 3 (t ) ) ; d 2 = d 2 (t ) =
d1 = d1 (t ) = ⋅ cos(θ 3 (t ) ). (A.14)
2 2
Dynamic equation of motion for Body 4 – Revolute joints C & D (Figure A.2.b):
m4 ⋅ ay4 = ∑Fyi ; ⇒ m4 ⋅ ay4 = 0 , (A.15)
m4 ⋅ az4 = ∑Fzi ; ⇒ m4 ⋅ az4 = FzC4 − FzD4 − m4 ⋅ g , (A.16)

+ CG
J c4 ⋅ ε 4 = ∑ M x4 ; ⇒ J c4 ⋅ ε 4 (t ) = 0 . (A.17)
when d1 and d2 are substitution by expressions:
- 156 -
CD CD
⋅ sin(θ 4 (t )) ; e2 = e2 (t ) =
e1 = e1 (t ) = ⋅ cos(θ 4 (t ) ) . (A.18)
2 2
Dynamic equation of motion for Body 5 – Revolute joints B & D (Figure A.2.c):
m5 ⋅ ay5 = ∑Fyi ; ⇒ m5 ⋅ ay5 = FyB − FyD6 , (A.19)
m5 ⋅ az5 = ∑Fzi ; ⇒ m5 ⋅ az5 = FzB + FzD4 − FzD6 − m5 ⋅ g , (A.20)

+CG
J c5 ⋅ ε 5 = ∑ M x5 ; ⇒ J c5 ⋅ ε 5 (t ) = FyB ⋅ f1 − FzB ⋅ f 2 + FyD6 ⋅ f1 − FzD6 ⋅ f 2 + FzD4 ⋅ f 2 . (A.21)
when e1 and e2 are substitution by expressions:

BD BD
f 1 = f 1 (t ) = ⋅ sin(θ 5 (t )) ; f 2 = f 2 (t ) = ⋅ cos(θ5 (t ) ) . (A.22)
2 2
a) FzD6 θ (t) b)
6
z7 FzH
7 H
z6 D FyD6
FyH
az6 M az7
6 C6 ε6(t) C7
ay6 y6 ay7 y7
ε7(t) FyI
m6⋅g
FW = FzM I
FyI I m7⋅g FzI
FzI Body 6 Body 7
Figure A.3: Free - body diagram for dynamic analysis of the; a) Body 6, and b) Body 7
Dynamic equation of motion for Body 6 – Revolute joints D & I (Figure A.3.a):
m6 ⋅ ay6 = ∑Fyi ; ⇒ m6 ⋅ ay6 = FyD6 − FyI , (A.23)
m6 ⋅ az6 = ∑Fzi ; ⇒ m6 ⋅ az6 = FzD6 − FzI − m6 ⋅ g , (A.24)

+ CG
J c6 ⋅ ε 6 = ∑M x6 ; ⇒ J c6 ⋅ ε 6 (t ) = FyI ⋅ g1 − FzI ⋅ g 2 + FyD6 ⋅ g1 − FzD6 ⋅ g2 . (A.25)
when f1 and f2 are substitution by expressions:

DI DI
⋅ sin(θ 6 (t ) ) ; g 2 = g 2 (t ) =
g1 = g1 (t ) = ⋅ cos(θ 6 (t ) ). (A.26)
2 2
Dynamic equation of motion for Body 7 – Revolute joints I & H (Figure A.3.b):
m7 ⋅ ay7 = ∑Fyi ; ⇒ m7 ⋅ ay7 = FyH + FyI , (A.27)
m7 ⋅ az7 = ∑Fzi ; ⇒ m7 ⋅ az7 = FzH + FzI + Fw − m7 ⋅ g , (A.28)
- 157 -
+ CG
J c7 ⋅ ε 7 = ∑M x7 ; ⇒ J c 7 ⋅ ε 7 (t ) = FyH ⋅ h1 − FzH ⋅ h2 + Fw ⋅ h2 − FyI ⋅ h1 − FzI ⋅ h2 . (A.29)
when f1 and f2 are substitution by expressions:
HI HM
h1 = ; h2 = (A.30)
2 2
So, we have 21 equations with 21 unknown. Suitable ways to solve these equations is
their presentation in the matrix form, presented below:
A(t ) ⋅ X (t ) = B (t ) (A.31)
For finding unknowns reaction forces X(t) it is required the equation (A.31) to transfer
in matrix form, given by expression:
X ( t ) = [ A( t ) ] ⋅ B ( t )
−1
(A.32)
Equation (A.32) enables to find the unknowns reaction forces that act in suspension
mechanism by using appropriate software such as MathCAD, MATLAB, etc. Elements in
matrix A(t) and B(t) are parameters which are defined and are in function of the ground
excitation by changing on the time. Therefore, unknown reactions forces are also in function
of the time.
1 0 0 0 −1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 1 0 0 0 −1 0 −1 0 0 0 0 0 0 0 0 0 0 0 0 0
 
b1 − b2 0 0 b1 − b2 0 − b2 0 0 0 0 0 0 0 0 0 0 0 0 0
 
0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 −1 0 0 0 0
0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 −1 0 0 0
 
0 0 0 0 c1 − c2 0 0 0 0 0 0 c3 − c4 0 0 c1 − c2 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 −1 0 1 0 0 0 0 0 0
 
0 0 0 0 0 0 0 0 0 0 0 0 0 −1 0 1 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 d1 d2 d1 d2 0 0 0 0 0
 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
A(t ) =  0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 0 0 0
 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 1 0 0 0 0 0 0 0 −1 0 0 0 0 0 0 0 0 0 0
 
0 0 0 1 0 0 0 0 0 1 0 −1 0 0 0 0 0 0 0 0 0
 
0 0 f1 − f2 0 0 0 0 0 f2 f1 − f2 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 −1 0 0
 
0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0
0 0 0 0 0 0 0 0 0 0 g1 − g2 0 0 0 0 0 0 g1 − g2 0
 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 1
 
 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 h1 − h2 − h1 − h2 h2 
- 158 -
 F yA (t )   m1 ⋅ a y1 (t ) 
 F (t )   m1 ⋅ a z1 (t ) + m1 ⋅ g 
 zA   
 F yB ( t )   J c1 ⋅ ε 1 ( t ) 
   
 FzB (t )   m2 ⋅ a y 2 (t ) − Fysd (t ) 
 F yC 2 (t )   m2 ⋅ a z 2 (t ) + m2 ⋅ g + Fzsd (t ) 
   
 FzC 2 ( t )   J c 2 ⋅ ε 2 (t ) − Fysd (t ) ⋅ b3 − Fzsd (t ) ⋅ b4 
 F yC 4 (t )   m 3 ⋅ a y 3 (t ) 
   
 FzC 4 ( t )   m3 ⋅ a z 3 (t ) + m3 ⋅ g 
 F (t )  J c 3 ⋅ ε 3 (t ) 
 yD 4   
 FzD 4 (t )   m 4 ⋅ a y 4 (t ) 
X (t ) =  F yD 6 (t )  B(t ) =  m 4 ⋅ a z 4 (t ) + m 4 ⋅ g 
   
 FzD 6 ( t )   J c 4 ⋅ ε 4 (t ) 
 F (t )   m5 ⋅ a y 5 ( t ) 
 yE   
 FzE (t )   m5 ⋅ a z 5 ( t ) + m 5 ⋅ g 
   
 F yG ( t )   J c5 ⋅ ε 5 ( t ) 
 FzG ( t )   m6 ⋅ a y 6 (t ) 
   
 F yH ( t )   m6 ⋅ a z 6 (t ) + m6 ⋅ g 
 FzH ( t )   J c 6 ⋅ ε 6 (t ) 
   
 F yI (t )   m7 ⋅ a y 7 (t ) 
 FzI (t )   m6 ⋅ a 7 (t ) + m7 ⋅ g 
   
 Fw (t )   J c 7 ⋅ ε 7 (t ) 
Determining of the projection of the reaction forces in x and y axis and total reaction
forces in each joint of the suspension mechanism is written by the following expression:
F j (t ) = (F ) + (F )
yj
2
zj
2
(A.33)
where” j” are joint of suspension mechanism such as A, B, C2, C4, D4, D6, E, F, G, H and I
joint. Solution of the unknowns’ reaction forces will be obtained in a function of time when
wheels of the vehicle moves from minimal (-250 mm) to maximal (+250 mm) values as a
consequence of the vertical wheel force act on the suspension mechanism. Other determined
forces which act in the suspension mechanism except reaction forces are spring forces,
damper forces and gravity forces of the mechanism linkages.
Following, in appropriate proportion are presented results of unknown reaction forces
for parameters given in table 5.6 through the relevant diagrams in function of the time.
- 159 -
24000
Fs - Spring force
22000 Fd - Damping force
Spring, damper, S&D and wheel forces [N] Fsd - Spr.&Damp. force
20000 Fw - Wheel force
18000 ks = 142 [N/mm]
16000 csh = 14.2 [Ns/mm]
14000
zmax = 250.00 [mm]
12000 t = π [s],
10000 Fw = 8461.00 [N]
8000
6000
4000 zrest = 0.0 [mm] zrest = 0.0 [mm]
2000 t = π/2 [s] t = 3π/2 [s]
0
-2000 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0
Time [s]
Figure A.4: Spring, damper, spring & damper and wheel forces as a function of time
16500
FyA - Force in y - axis
15000 FzA - Force in z - axis
FA - Total force
13500
Reaction forces; FyA, FzA and FA [N]
12000
10500
9000
7500
6000
4500
3000
1500
0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0
Time [s]
Figure A.5: Reaction forces in the joint A as a function of time
- 160 -
16500
FyC2 - Force in y - axis
13500 FzC2 - Force in z - axis
FC2 - Total force
Reaction forces; FyC2, FzC2 & FC2 [N]
10500
7500
4500
1500
-1500 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0
-4500
-7500
-10500
-13500
Time [s]
Figure A.6: Reaction forces in the joint C2 as a function of the time
16500
FyB - Force in y - axis
15000 FzB - Force in z - axis
FB - Total force
13500
Reaction forces; FyB, FzB and FB [N]
12000
10500
9000
7500
6000
4500
3000
1500
0
-1500 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0
-3000
Time [s]
Figure A.7: Reaction forces in the joint B as a function of time
- 161 -
16500
FyD6 - Force in y - axis
15000 FzD6 - Force in z - axis
FD6 - Total force
13500
Reaction forces; FyD6, FzD6 and FD6 [N]
12000
10500
9000
7500
6000
4500
3000
1500
0
-1500 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0
-3000
Time [s]
Figure A.8: Reaction forces in the joint D6 as a function of time
16500
FyC4 - Force in y - axis
15000 FzC4 - Force in z - axis
FC4 - Total force
13500
Reaction forces; FyC4, FzC4 and FC4 [N]
12000
10500
9000
7500
6000
4500
3000
1500
0
-1500 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0
-3000
Time [s]
Figure A.9: Reaction forces in the joint C4 as a function of time
- 162 -
16500
FyD4 - Force in y - axis
15000 FzD4 - Force in z - axis
FD4 - Total force
13500
Reaction forces; FyD4, FzD4 and FD4 [N]
12000
10500
9000
7500
6000
4500
3000
1500
0
-1500 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0
-3000
Time [s]
Figure A.10: Reaction forces in the joint D4 as a function of time
27000
FyE - Force in y - axis
FzE - Force in z - axis
24000 FE - Total force
Reaction forces; FyE, FzE and FE [N]
21000
18000
15000
12000
9000
6000
3000
0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0
Time [s]
Figure A.11: Reaction forces in the joint E as a function of the time
- 163 -
16500
FyG - Force in y - axis
15000 FzG - Force in z - axis
Reaction forces; FyG, FzG and FG [N] 13500 FG - Total force
12000
10500
9000
7500
6000
4500
3000
1500
0
-1500 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0
-3000
-4500
-6000
-7500
-9000
Time [s]
Figure A.12: Reaction forces in the joint G as a function of time
16500
FyI - Force in y - axis
15000 FzI - Force in z - axis
FI - Total force
13500
Reaction forces; FyI, FzI and FI [N]
12000
10500
9000
7500
6000
4500
3000
1500
0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0
-1500
-3000
Time [s]
Figure A.13: Reaction forces in the joint I as a function of the time
- 164 -
16500
FyH - Force in y - axis
15000 FzH - Force in z - axis
13500 FH - Total force
12000
Reaction forces; FyH, FzH and FE [N]
10500
9000
7500
6000
4500
3000
1500
0
-1500 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0
-3000
-4500
-6000
-7500
-9000
Time [s]
Figure A.14: Reaction forces in the joint H as a function of the time
- 165 -
Appendix B: Simulation and Optimization of the Suspension Parameters

for Passive, Semi-active and Active System for a Terrain Vehicle
Figure B.1: Parts of the MATLAB codes for solving differential equations of motion for
passive non-linear suspension system
- 166 -
Figure B.2: Part of MATLAB codes for solving differential equations of motion for non-
linear semi-active suspension system
- 167 -
Figure B.3: Configuration MATLAB/Simulink parameters for solving the differential

equations of motion
ksmin ... ktmin ... ksmax ... ktmax ... ksi ... kti ...
σ &z& σ F ksop cshop ktop msop muop

s t
Figure B.4: Window of Multi Objective Genetic Algorithms performed in MATLAB

Optimization Tool
- 168 -
Start
Limits of the design variables

Design ksmin≤ks≤ksmax; cshmin≤csh≤cshmax Ground excitation, zr(t)
limits ktmin≤kt≤ktmax; msmin≤ms≤msmax
mumin≤mu≤mumax
MOGAs process
Initial population of the design

Step 1# variables (ksi, cshi, kti, msi, mui)
Fitness function
Dynamic equations of (Objective function)
Step 2# motion: (), (), z(t) and
Selection function
Step 3# (Tournament)
Reproduction
Step 4#
Crossover
Improved design
NO Stopping variables:
Step 5# Mutation
criteria *
ks, *csh, *kt, *ms, *mu
YES
Optimal design variables:

ksop, cshop, ktop, msop, muop
Figure B.5: Flowchart diagram of the optimization process utilized in MOGAs
- 169 -
Appendix C: Design of the Developed Terrain Vehicle
Figures C.1 introduces design of the terrain vehicle in 2D/3D with some main geometric
dimension performed in Pro/ENGINEER environment.
CG
D/2
L lF
w
lR hCG
α
1900
2100
2800
Figure C.1: 2D/3D modeling of the terrain vehicle with four wheels drive
and four wheels steering performed in Pro/ENGINEER environment [6]
- 170 -
Figures C.2 and Figure C.3 3D present design of the front and rear mechanism and tires
for terrain vehicle, performed
med in Pro/ENGINEER environment.
environment
Figure C.2: 3D modeling of a front and rear suspension mechanism,

engine and driver perfor
performed in Pro/ENGINEER environment [6]
C.3 3D modeling of a front and rear wheels

Figure C.3:
and driver perfor
performed in Pro/ENGINEER environment [6]
- 171 -
Appendix D: Bibliography of the Candidate
Personal bibliography for the period 2003 - 2013
ARTICLES AND OTHER COMPONENT PARTS
1.01 Original scientific article
[1] Lajqi Shpetim, Pehan Stanislav, Lajqi Naser, Gjelaj Afrim, Pšeničnik Jože, Sašo
Emin. Design of Independent Suspension Mechanism for a Terrain Vehicle with Four
Wheels Drive and Four Wheels Steering. Annals of Faculty Engineering Hunedoara –
International Journal of Engineering (2013), Fascicule1, Tome XI 2013, pp. 101-108.
http://annals.fih.upt.ro/pdf-full/2013/ANNALS-2013-1-12.pdf.
[2] Lajqi Shpetim, Gugler Jürgen, Lajqi Naser, Shala Ahmet, Likaj Ramë: Possibilities
Experimental Method to Determine the Suspension Parameters in a Simplified Model of
a Passenger Car. International Journal of Automotive Technology (2012), vol. 13, no.
4, pp. 615-621. http://link.springer.com/article/10.1007%2Fs12239-012-0059-7.
[3] Lajqi Shpetim, Stanislav Pehan. Designs and Optimizations of Active and Semi-
Active Non-linear Suspension System for a Terrain Vehicle. Strojniški vestnik -
Journal of Mechanical Engineering (2012), vol. 58, no.12, pp. 732-743.
DOI:10.5545/sv-jme.2012.776. http://www.sv-jme.eu/archive/sv-jme-volume-2012/sv-
jme-58-12-2012.
[4] Lajqi Shpetim, Pehan Stanislav, Lajqi Naser, Baxhaku Bashkim, Ymeri Besim.
Influences of the Suspension Parameters on the Vehicle Suspension Performance for a
Terrain Vehicle. Journal of Mechanics Engineering and Automation (2012), New York,
vol. 2, no. 9, pp. 550-554. http://www.davidpublishing.com/show.html?8218.
[5] Baxhaku Bashkim, Tytyri Hajredin, Lajqi Shpetim (2012): One of Scenarios for
Reduction the Emission of Pollutant for Motor Vehicle in the Territory of the Republic
of Kosovo, Journal of Mechanics Engineering and Automation (2012), New York,
[6] Mehmeti Xhemajl, Lajqi Naser, Baxhaku Bashkim, Lajqi Shpetim, Tytyri Hajredin.
Preventive Maintenance of Passengers Cars Driving in the Territory of the Republic of
- 172 -
Kosovo. Journal of Mechanics Engineering and Automation (2012), New York,

[7] Lajqi Naser, Lajqi Shpetim, Vataj Gjelosh, Avdiu Arben. The influence of Cavitations
Phenomenon and Water Hammer in Pumping Station of Mitrovica. Journal of Institute
Alb-Shkenca – AKTET, ISSN 2073-2244 (2011), vol. 4, no. 2, pp. 359-365.
http://alpa.mali-it.eu/journal/aktet//vol/?l=en.
[8] Lajqi Shpetim, Baxhaku Bashkim, Lajqi Naser. Influence of Planetary Gear in
Improving Efficiency of Continuous Variable Transmission in Passenger Vehicle.
Journal of Institute Alb-Shkenca - AKTET, ISSN 2073-2244 (2011), Tirana, vol. 4, no.
2, pp. 219-226. http://alpa.mali-it.eu/journal/aktet//vol/?l=en.
1.04 Professional article
[9] Mehmeti Xhemail, Lajqi Naser, Lajqi Shpetim. Developing of the Model of the
Functioning Vehicle Maintenance Which is Driven in Territory of the Republic of
Kosova. Annual Conference of Science - Science Week (2012), Prishtina, Kosova.
[10] Lajqi Shpetim, Elezaj Shaqir, Lajqi Naser. Possibility of Installing Micro Hydro –
Generator in Territories with Hydraulic Potential of Water. 5th Conference of Albanian
Association of Thermotechnics (2010), Prishtina, Kosova.
[11] Baxhaku Bashkim, Lajqi Naser, Lajqi Shpetim. Definitions of the Reaction in Main
Support Bearing for Motor Perkins P-4 Massey-Ferguson. Symposium of Mechanical
Engineering (2004), Prishtina, Kosova.
[12] Fejzullahu Xhemajl, Lajqi Shpetim, Zhitia Ilir, Vataj Gjelosh. Analyzing of the
Cavitations Phenomena in Main Pumping Station in Batllava. Professional and
Scientific Magazine (2003), Ferizaj, Kosova.
1.08 Published scientific conference contribution
[13] Lajqi Shpetim, Pehan Stanislav, Lajqi Naser, Gjelaj Afrim, Pšeničnik Jože, Emin Saso.
Design of Independent Suspension Mechanism for Terrain Vehicle with Four Wheels
Drive and Four Wheel Steering. The 4th International Scientific Conference,
Management of Technology Step to Sustainable Production (2012), Zader, Croatia
MOTSP 2012 Conference proceeding, pp. 230-237.
- 173 -
[14] Lajqi Shpetim, Pehan Stanislav, Pšeničnik Jože, Lajqi Naser. Tracking the Prescribed
Path Through an Autonomous Mobile Robot with Four Wheel Steering. The 3th
International Scientific Conference, Management of Technology Step to Sustainable
Production (2011), Bol, Croatia, MOTSP 2011 Conference proceedings, pp. 94-101.
[15] Pehan Stanislav, Lajqi Shpetim, Pšeničnik Jože, Flašker Jože. Modeling and Simulation
of off Road Vehicle with Four Wheel Steering. The 7th International Scientific
Conference, Research and Development of Mechanical Elements and Systems (2011),
IRMES 2011, Conference Proceedings, pp. 77-83.
[16] Baxhaku Bashkim, Lajqi Naser, Lajqi Shpetim, Tytyri Hajredin. Possibility of
Reduction Emissions of Pollutants in Urban Areas. 1st International Scientific
Conference on Engineering Manufacturing and Advanced Technologies (2010), Mostar,
Bosnia and Herzegovina, MAT 2010 Conference proceedings.
[17] Baxhaku Bashkim, Lajqi Naser, Cakolli Heset, Lajqi Shpetim. Influence of motor
vehicle in pollutant emission in urban environments. 14th International Research/Expert
Conference, Trends in the Development of Machinery and Associated Technology
(2010), Mediterranean Cruise, TMT 2010 Conference proceedings, pp. 301 – 304.
http://www.tmt.unze.ba/proceedings2010.php.
[18] Baxhaku Bashkim, Lajqi Naser, Lajqi Shpetim. Ecology and road transport in Republic
of Kosova. 14th International Research/Expert Conference, Trends in the Development
of Machinery and Associated Technology (2010), Mediterranean Cruise, TMT 2010
Conference proceedings, pp. 313 – 316. http://www.tmt.unze.ba/proceedings2010.php.
[19] Lajqi Shpetim, Baxhaku Bashkim, Lajqi Naser. Influence of planetary gear in
improving efficiency of continuous variable transmission in passenger vehicle.
International Annual Meeting of Alb-Science Institute (2010), Tirana, Albania, pp. 462.
[20] Lajqi Naser, Lajqi Shpetim, Vataj Gjelosh, Avdiu Arben. Influence of N.P.S.H. and
water hammer in pumping station in Mitrovica. International Annual Meeting of Alb-
Science Institute (2010), Tirana, Albania, pp. 463.
[21] Lajqi Shpetim, Jürgen Gugler, Lajqi Naser, Shala Ahmet, Lika Ramë. Experimental
Possibilities for Determination of Suspension Parameters by the Simplified Model of
Passenger Car. 2nd International Scientific Conference, Management of Technology Step
to Sustainable Production (2010), Rovinj, Croatia. MOTSP 2010 Conference
proceedings.
[22] Lajqi Naser, Baxhaku Bashkim, Lajqi Shpetim. Effect of intercooler on turbocharged
diesel engine performance. 13th International Research/Expert Conference Trends in the
- 174 -
Development of Machinery and Associated Technologies (2009), Hammamet, Tunisia.

TMT 2009 Conference proceedings, pp. 437 – 440.
[23] Lajqi Shpetim, Ibrahimi Nijazi, Baxhaku Bashkim, Lajqi Naser. Analyses of Powertrain
System in a Passenger Vehicle with Using Planetary Gear Combine with Continuously
Variable Transmission. 13th International Research/Expert Conference Trends in the
Development of Machinery and Associated Technologies, 16-21 October 2009,
Hammamet, Tunisia. TMT 2009 Conference proceedings, pp. 641 – 644.
[24] Lajqi Shpetim, Ibrahimi Nijazi, Baxhaku Bashkim, Lajqi Naser. Modelling of
Continuous Variable Power Split Transmission in Hybrid Electric Vehicle. 6th
Research/Expert Conference with International Participations (2009), Neum, Bosnia
and Herzegovina. QUALITY 2009 Conference proceedings, pp. 443 – 448.
http://www.quality.unze.ba/zbornici/QUALITY%202009/index.php.
[25] Lajqi Naser, Baxhaku Bashkim, Lajqi Shpetim. Model of Coupling Turbocharger and
Diesel Engine. 12th International Research/Expert Conference Trends in the
Development of Machinery and Associated Technologies (2008), Istanbul, Turkey. TMT
2008 Conference proceedings, pp. 685 – 688.
[26] Lajqi Naser, Baxhaku Bashkim, Lajqi Shpetim. Model of Calculation of Characteristic
Parameters for Marine Motors. 11th International Research/Expert Conference Trends
in the Development of Machinery and Associated Technology (2007), Hammamet,
Tunisia. TMT 2007 Conference proceedings, pp. 1187 – 1190.
MONOGRAPHS AND OTHER COMPLETED WORKS
2.05 Other educational material
[27] Lajqi Shpetim, Ramadani Riad. Technical drawing with Descriptive Geometry. Material
for exercise, Faculty of Mechanical Engineering, 2010, Prishtina, Kosova.
[28] Lajqi Shpetim. Power transmitter. Material for exercise, Faculty of Mechanical
Enginiering, 2010, Prishtina, Kosova.
- 175 -
2.09 Master's thesis
[29] Lajqi Shpetim. Planetary Transmissions Combined with Continuously Variable

Transmission in Hybrid Electric Vehicle. University of Prishtina, Faculty of Mechanical
Engineering (2008), Master Thesis, Prishtina, Kosova, pp. 149.
2.11 Diploma thesis
[30] Lajqi Shpetim. Determining of the reactions and dimensioning of the crank shaft for
Perkins Engine P-4 Massey-Ferguson, University of Prishtina, Faculty of Mechanical
Engineering (2002), Diploma Thesis, Prishtina, Kosova, pp. 150.
2.14 Project documentation (feasibility study, preliminary design, detail design)
[31] Lajqi Shpetim, Vataj Gjelosh. Detail design of extension of water treatment plant in
Mitrovica for water company "MITROVICA" sh.p.k., K.K. Mitrovicë, 2010.
[32] Lajqi Shpetim, Vataj Gjelosh. Geomechanical & hydraulic study for alternate routing
of slag granulation return channel, for enterprise "NewCo FeroNickel L.L.C." sh.p.k.,
K.K. Drenas, 2010.
[33] Bajraktari Musli, Lajqi Naser, Lajqi Shpetim, Vataj Gjelosh. Elaborate for periodically
inspection and attesting of crane and elevators for enterprise "NewCo FeroNickel
L.L.C." sh.p.k., K.K. Drenas, 2006-2010.
[34] Bajraktari Musli, Lajqi Naser, Lajqi Shpetim, Vataj Gjelosh. Detail design of technical
– technologic project for dryer system for mineral production from Benard Company in
Italy. Financing from enterprise "NewCo FeroNickel L.L.C." sh.p.k., K.K. Drenas,
2010.
– technologic project for rock crusher & mobile separation of sand for private enterprise
"Alko-Impex-General Construction" sh.p.k. in Vill. Dabishevc, K.K. Prishtinë,
2009.
– technologic project for mobile crusher for private enterprise "I. SEFERI", Suharekë,
2006.
- 176 -
– technologic project for base separation of sand for private enterprise "BENITA
COMPANY", Klinë, 2005.
– technologic project for concrete base for private enterprise "BENITA COMPANY",
Klinë, 2005.
– technologic project for concrete base for private enterprise "GRANITI" Istog, 2005.
– technologic project for asphalt base for private enterprise "GRANITI" Istog, 2005.
….
– technologic project for asphalt base for private enterprise "DRINI COMPANY",
Malishevë, 2005.
- 177 -
Appendix E: Curriculum Vitae
Name and surname: Shpetim LAJQI

Date and place of birth: 11.04.1977, Prapaqan
Permanent address: Rr. Rexhep Krasniqi p.n.
Lagja e Emshirit, Kulla 5, kati 3/3
10000 Prishtinë, Republika e Kosovës
Nationality: Kosovar
E-mail address: shpetim.lajqi@uni-pr.edu
Mobile phone: + 377 44 259 771
Driving licence: B1, B, C1, C, BE, C1E, CE, M, L, T
Education:
Postgraduates study:
Doctoral study: University of Maribor, Faculty of Mechanical Engineering, Maribor
Program/Institute: Mechanical Engineering/Institute for Structure and Machine Design
Date (from – to): 2010/11 – onward
Academic degree: Doctor of Science
Master study: University of Prishtina, Faculty of Mechanical Engineering, Prishtina
Program/Department: Mechanical Engineering/Department for Construction and Design
Title of thesis: Planetary Transmissions Combined with Continuously Variable
Transmission in Hybrid Electric Vehicles
Date (from – to): 2003/04 – 2007/08
Academic degree: Magister of Technical Science
Average score: 9.77
Graduate study: University of Prishtina, Faculty of Mechanical Engineering, Prishtina
Program/Department: Mechanical Engineering/Depart. Mech. Construction and Mechanism
Title of thesis: Determination of the Reactions and Dimensioning of the Crank Shaft
for Perkins Engine P-4 Massey - Ferguson
Date (from – to): 1995/96 – 2001/02
Academic degree: Graduate Mechanical Engineer
Average score: 9.20
- 178 -
Secondary school: Gymnasium “QAMO Vëllezërit Frashëri”, Deçan, Kosova

Date (from – to): 1991/92 – 1994/95
Degree: Diploma
Primary school: Primary school “Jusuf Gërvalla”, Prapaqan, Deçan, Kosova
Date (from – to): 1983/84 – 1990/91
Awards and Scholarships:

Awards: Diploma as the best student, issued from University of Prishtina
Diploma as the best student, issued from Faculty of Mechanical
Engineering, University of Prishtina,
Scholarship: Slovene human resource development and scholarship funds, Slovenia
JoinEU SEE, Austria
Austrian Exchange Service, Austria
University of Prishtina, Kosova
Municipality of Deçan, Kosova
Employment:
Institution: University of Prishtina, Faculty of Mechanical Engineering
Academic title: Assistant
Date (from – to): 2011– onward
Academic title: New Assistant
Date (from – to): 2007– 2011
Academic title: Teaching Assistant (collaborator)
Date (from – to): 2003 – 2007
Company: Regional Water Company “PRISHTINA”
Positions: Assistant Director, Maintenance Manager, Engineer, etc.
Date (from – to): 2002 – 2007
- 179 -
Training and some other activities:

Date (from – to) Description of training and other activities
06 – 28 November 2007 Research in Austrian Universities, Graz University of
Technology/Faculty of Mechanical Engineering, Vehicle Safety
Institute, Graz, Austria,
15.10.07 – 02.11.07 Business Skills Training organized by the Business Start – Up
03.12.07 – 14.12.07 Centre Kosovo, Prishtina, Kosova,
16 – 20 May 2004 Regional Seminar from South - East Europe for preparation
projects, Ohrid, Macedonia,
09 – 10 Mars 2004 Program for trainings of management,
17 December 2003 Investment plan and Business Plan,
19 – 26 May 2003 Study visit in “Helsinki Water Company” and “Tallinn Water
Company”, Finland and Estonia.
24 – 25 April 2003 Study visit in “Water Company Essen”, Essen, Germany
- 180 -
UNIVERSITY OF MARIBOR
FACULTY OF MECHANICAL ENGINEERING
DECLARATION OF THE DOCTORAL CANDIDATE
LAJQI M. Sc., B. Sc. Mech. Eng. with enrolling nr. 95031843

The undersigned: Shpetim LAJQI,
declare that,
the doctoral thesis entitled ““Suspension

Suspension and Steering System Development of a Four
Wheel Drive and Four Wheel Steered Terrain Vehicle”
Vehicle
• present our results of the research works,

• submitted thesis in whole or part has not been submitted into any kind of academic
programs in other faculties or universities,
• presented results are correctly specified, and
• I didn’t breach copyright and other intellectual property.
Maribor, 04.03.2013 Signature: ___________________________

4WS System

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

4WS System

Uploaded by

Copyright:

Available Formats

UNIVERSITY OF MARIBOR

FACULTY OF MECHANICAL ENGINEERING

SUSPENSION AND STEERING SYSTEM

March, 2013 M. Sc. Shpetim LAJQI

SUSPENSION AND STEERING SYSTEM

March, 2013 Author: M. Sc. Shpetim LAJQI

Author: Shpetim LAJQI, M. Sc., B. Sc. Mech. Eng.

I would like to express my gratitude and appreciation to my

At the same time I would like to express my thanks to

For financial support my thanks go to the Slovene Human

Further thanks go to my parents and family for their dedication,

2 LITERATURE STUDY ........................................................................................... 8

3 RESEARCH METHODOLOGY ........................................................ 25

4 DEVELOPMENT OF THE SUSPENSION SYSTEM......................... 33

4.2 Geometric parameters involved in the suspension system ...................................... 37

5 MODELING, SIMULATION AND OPTIMIZATION OF THE

6 DESIGNS AND OPTIMIZATIONS OF PASSIVE, ACTIVE AND

6.5 Simulation of the vehicle’s suspension by using MATLAB/Simulink .................. 107

7 DEVELOPMENT OF THE FOUR WHEEL STEERING SYSTEM

8 SUMMARY AND CONCLUSION ................................................... 142

Figure 2.1: Coach, constructed around 1650 years, exhibited in Automobile

Figure A.1: Free-body diagram; a) Body 1, and b) Body 2 ............................................... 155

SUSPENSION AND STEERING SYSTEM DEVELOPMENT OF A FOUR

RAZVOJ OBES IN KRMILNEGA SISTEMA TERENSKEGA VOZILA S

Cilj doktorske disertacije je zasnovati, oblikovati, optimizirati in analizirati sistem obes in

LIST OF USED SYMBOLS

Symbol Unit Explanation of the used symbol

ks1 [Nm-1] Spring stiffness non-linear coefficient

R4WS [mm] Vehicle turning radius at the vehicle center gravity

zs [m] Vertical displacement of the sprung mass

θi(θ3) [deg] Angular displacements of the ith suspension mechanism links

LIST OF USED ABBREVIATIONS

Abbreviations Explanation of the used abbreviations

CAD Computer Aided Design

1.2 Problem definition

No. Characteristics of a new terrain vehicle Value Unit Remarks

1.2.1 Methodology for examining problem

1.3 Expected original scientific contribution

1.4 Research Objectives

The objectives of this research were focused as follows:

1.5 Structure of the doctoral thesis

The Doctoral thesis is organized into eight chapters.

2.1 Historical evolution

Figure 2.1: Coach, constructed around

Solid axle Leaf spring

Steering by rotate of Steering by rotaion

Rear solid axis ICE Tiller

2.2 Function and classification of the suspension system

2.2.1 The solid axle suspension system

Solid axle tilts Tire

2.2.2 The independent suspension system

The disadvantages of the independent suspension system are considered to be the

Figure 2.8: Design of the independent suspension system;

2.2.3 The controllability strategies for the damping system

2.3 Functions and classifications of the steering systems

Pinion gear Rack shaft

Kingpin axis Steering linkages

Figure 2.10: Mechanical front wheels steering system

2.3.1 Power steering system

Figure 2.11: Hydraulic power steering Figure 2.12: Electric amplifier

2.4 Four wheels steering systems

2.4.1 Mechanical four wheels steering system

Figure 2.13: Mechanical four wheels steering system

2.4.2 Hydraulic four wheels steering system

introduces vector position R1 which is the output link with